- Email: [email protected]
Differential regulation of cyclooxygenase-2 in the rat hippocampus after cerebral ischemia and ischemic tolerance
Neuroscience Letters 393 (2006) 231–236
Differential regulation of cyclooxygenase-2 in the rat hippocampus after cerebral ischemia and ischemic tolerance Jeong-Sun Choi, Ha-Young Kim, Myung-Hoon Chun, Jin-Woong Chung, Mun-Yong Lee ∗ Department of Anatomy, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Socho-gu, Seoul 137-701, Republic of Korea Received 25 July 2005; received in revised form 14 September 2005; accepted 28 September 2005
Abstract We investigated the temporal changes and cellular localization of cyclooxygenase-2 (COX-2) in the rat hippocampus during the induction of acquired ischemic tolerance by sublethal ischemia, and compared these changes with those occurring following transient forebrain ischemia. Adult male Sprague Dawley rats were subjected to either 10 min of lethal global ischemia with or without 3 min of sublethal ischemic preconditioning, or 3 min of ischemia only. A short (3 min) cerebral ischemia as well as lethal ischemia with preconditioning substantially and significantly upregulated COX-2 expression in dentate granule cells, as confirmed by immunoblot analysis. This became evident by 4 h, peaked at 1–3 days, and returned to the basal level around 7 days. COX-2 expression was also increased in CA2 and CA3 neurons, although with weaker staining intensity, but in CA1 neurons very weak immunoreactivity was transiently observed. In the ischemic hippocampus, however, in agreement with previous reports, COX-2 expression was induced strongly in vulnerable CA1 and hilar neurons as well as in resistant CA3 and dentate granule cells. These data demonstrated that COX-2 expression is upregulated in neuronal subpopulations destined to survive, i.e., in CA3 and dentate granule cells after ischemia and ischemia-tolerance induction, as well as in ischemia-vulnerable neurons, i.e., in CA1 neurons after lethal ischemia, suggesting that hippocampal neuronal subpopulations have differential sensitivity to COX-2 upregulation. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: COX-2; Hippocampus; Ischemic preconditioning; Dentate granule cells
Cyclooxygenase (COX), a rate-limiting enzyme in the synthesis of prostaglandins (PG) from arachidonic acid, produces PGH2, which in subsequent steps gives rise to PGs with various physiological functions [6,9,20]. COX-2, the inducible form, is thought to be involved in mechanisms of ischemic brain injury. Cerebral ischemia upregulates COX-2 in neurons, glial cells and infiltrating leukocytes in injured brain [8,13,15]. In addition, pharmacological inhibition of COX-2 and genetic deletion of COX-2 reduce infarct size, and neuronal overexpression of COX-2 increases cerebral infarction [4,5,14,15]. These observations suggest that COX-2 plays a deleterious role in ischemic injury. Recently, Horiguchi et al. [7] reported that ischemic tolerance by cortical spreading depression induced immediate upregulation of COX-2. In addition, COX-2 is a critical mediator of myocardial protection by ischemic preconditioning [1]. Although upregulation of COX-2 in the ischemic hippocampus
∗
Corresponding author. Tel.: +82 2 590 1108; fax: +82 2 536 3110. E-mail address: [email protected] (M.-Y. Lee).
0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.09.074
is attenuated by induction of ischemic tolerance using real-time polymerase chain reaction (RT-PCR) [3], little information is known about the selective expression and regulation of COX-2 after a preconditioning ischemic stimulus. A brief episode of sublethal ischemia is known to provide resistance against damage induced by a subsequent lethal ischemia [10,16]. This study was therefore performed to investigate the temporal changes and cellular localization of COX-2 in the rat hippocampus during the induction of acquired ischemic tolerance by sublethal ischemia. We also compared these changes with those occurring following transient forebrain ischemia. Transient forebrain ischemia was induced by four-vessel occlusion and reperfusion, as previously described [17] with minor modifications [2]. Briefly, the vertebral arteries were electrocauterized and cut completely to stop circulation in these vessels. After 24 h, both common carotid arteries were occluded for 3 min (sublethal ischemia) or 10 min (lethal ischemia) using miniature aneurismal clips. In some experiments, a 3-min period of occlusion (ischemic preconditioning) or sham operation was followed by 3 days of reperfusion, after which 10 min of occlusion was induced again. Only animals with complete EEG
232
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
flattening upon vascular occlusion were classified as ischemic and used in the study. Rectal temperature was maintained at 37.5 ± 0.3 ◦ C using a heating lamp during and after ischemia. Sham-operated rats with cauterized vertebral arteries and ligatures placed around the carotid arteries were used as controls. No animal convulsed or died following reperfusion or sham operation. Animals were sacrificed 4 and 12 h, and 1, 3, 7, and 14 days after reperfusion. For each time point, three rats were used for immunoblot analysis and seven for immunohistochemistry. Sham-operated animals were treated using the same schedule as the ischemic-reperfused animals. At each time point after reperfusion, animals were deeply anesthetized with 16.9% urethane (10 mL/kg) and killed by transcardial perfusion with a fixative containing 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) or by decapitation followed by dissection of the hippocampi, which were quickly frozen in liquid nitrogen. All experimental procedures performed on the animals were conducted with the approval of the Catholic Ethics Committee of the Catholic University of Korea, and were in accordance with the National Institute of Health’s Guide for the Care and Use
of Laboratory Animals (NIH Publication No. 80-23, revised 1996). For immunoblot analysis, coronal slices (400 m thick) were cut at the level of the septal hippocampus with a McIlwain tissue chopper. The dorsal hippocampal CA1 sectors and dentate gyrus regions were selectively dissected from the slices under microscopic observation, and were homogenized in ice-cold RIPA buffer (50 mM Tris buffer, pH 8.0; 150 mM NaCl; 1% NP-40; 0.5% deoxycholate; and 0.1% SDS). Equal amounts (20 g) of total protein from the CA1 sector or dentate gyrus of the hippocampi from sham-operated rats and rats that had undergone 3 or 10 min of ischemia were separated in 8% SDS–polyacrylamide gel electrophoresis (PAGE) and blotted onto cellulose membranes. Immunostaining of the blots was performed using a goat polyclonal antibody against COX-2 (Santa Cruz Biotechnology, Santa Cruz, CA, USA; diluted 1:1000). The immunoreactive bands were detected by enhanced chemiluminescence (ECL; Amersham, Arlington Heights, IL, USA). Densitometric analysis was performed using the Eagle Eye TMII Still Video System (Stratagene, La Jolla, CA, USA). Three
Fig. 1. Comparison of COX-2 expression in the rat hippocampus between ischemic tolerance (B, C, E and G) and cerebral ischemia (D, F and H). (A) In control sections of the hippocampus, weak COX-2 immunoreactivity was localized to neurons of the pyramidal cell and granule cell layers. DG, the dentate gyrus. (B) Four hours after 3 min of ischemia, COX-2 immunoreactivity increased significantly in the granule cells and slightly in CA2 and CA3 neurons. (C) One day after 3 min of ischemia, increased immunoreactivity in the granule cells was more evident and weakly labeled profiles had appeared in the pyramidal cell layers along CA1–CA3. (D) One day after 10 min of ischemia, COX-2 immunoreactivity had increased preferentially in the CA1 and dentate hilar region and also in granule cells and CA2 and CA3 neurons. (E–H) Higher magnification views of the boxed areas in (C) and (D), respectively. GL, the granule cell layer; pcl, the pyramidal cell layer; PL, the dentate hilar region; sr, the stratum radiatum. Scale bar = 300 m for (A–D); 50 m for (E–F); 100 m for (G–H).
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
animals were used for immunoblotting at each time point, and relative optical densities of the protein bands were obtained from three independent experiments each performed in triplicate. The data were expressed as mean ± S.E.M. Statistical significance was analyzed by analysis of variance (ANOVA) followed by Dunnett’s t-test; p < 0.01 was regarded as significant. For COX-2 immunohistochemistry, free-floating, 25-mthick sections were processed. After blocking with 10% normal donkey serum for 1 h, the sections were incubated overnight at 4 ◦ C with a goat polyclonal antibody against COX-2 (Santa Cruz Biotechnology; diluted 1:100). Primary antibody binding was visualized using peroxidase-labeled donkey anti-goat antibody (Jackson ImmunoResearch, West Grove, PA, USA; diluted 1:200), and 0.05% 3,3 -diaminobenzidine tetrahydrochloride and 0.01% H2 O2 as the substrate. The specificity of COX-2 immunoreactivities was confirmed by the absence of immunohistochemical reaction in sections from which the primary antibody was omitted or substituted with non-specific rabbit IgG. Histological staining with Cresyl violet was performed on sections adjacent to those processed for immunohistochemistry.
233
The distribution and cellular localization of COX-2 immunoreactivity in the hippocampus following 3 or 10 min of ischemia were examined using immunohistochemistry. A low level of basal COX-2 immunoreactivity occurred in neurons of the pyramidal cell and granule cell layers in the hippocampus of control rats at all time points (Fig. 1A). By 4 h after 3 min of ischemia, COX-2 immunoreactivity was significantly increased in dentate granule cells but only slightly increased in CA2 and CA3 neurons (Fig. 1B). By 1 day after 3 min of ischemia, COX2 immunoreactivity was progressively enhanced in granule cells and also more evident in CA2 and CA3 neurons (Fig. 1C and G). In addition, more weakly labeled profiles had appeared in the pyramidal cell layers in the CA1 region (Fig. 1E). In the ischemic hippocampus after 10 min of ischemia, COX-2 expression was upregulated in all hippocampal neurons after 1 day of reperfusion (Fig. 1D). The signals in CA1 neurons and the dentate hilar neurons were stronger than those of the dentate granule cells (Fig. 1F and H). Three days after 3 min of ischemia, increased immunoreactivity was still observed in dentate granule cells and in CA2 and
Fig. 2. Comparison of COX-2 expression in the rat hippocampus between ischemic tolerance (A, C, E and G) and cerebral ischemia (B, D, F and H). (A) Three days after 3 min of ischemia, increased COX-2 immunoreactivity was observed in granule cells and in CA2 and CA3 neurons. (B) Three days after 10 min of ischemia, intense immunoreactivity was observed in granule cells and CA3 neurons, and also in scattered surviving neurons in the CA1 region. (C) Seven days after the 3 min of ischemia, COX-2 immunoreactivity had returned toward control levels. (D) Seven days after the 10 min ischemic insult, neuronal COX-2 expression had returned to the control level, but immunoreactivity was observed in small cells showing the morphology of astrocytes, which were sparsely distributed in the strata radiatum and oriens of the CA1 region. (E–H) Higher magnification views of the boxed areas in (A–D). pcl, the pyramidal cell layer; sr, the stratum radiatum. Scale bar = 300 m for (A–D); 50 m for (E–H).
234
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
CA3 neurons, whereas no significant labeling was observed in preserved CA1 neurons (Fig. 2A and E). Three days after 10 min of ischemia, intense COX-2 immunoreactivity was observed in CA3 and granule cells, and in a few scattered surviving neurons in the CA1 region where most neuronal signals had disappeared (Fig. 2B and F). Preconditioned hippocampus that had been subjected to a subsequent 10 min of ischemia exhibited a labeling pattern similar to that seen in the hippocampus following 3 min of ischemia (data not shown). COX-2 immunoreactivity had returned toward control levels in 7 days (Fig. 2C and G) and 14 days (data not shown) after 3 min of ischemia. Seven days after 10 min of ischemia, the expression of COX-2 had returned to the control level, with a few labeled cells that had the morphology of astrocytes observed in the strata radiatum and oriens of the CA1 region (Fig. 2D and H), and the labeling pattern remained unchanged for 14 days (data not shown). The results of the immunoblot analysis after 3 or 10 min of ischemia were consistent with the immunohistochemical data. Immunoblot analyses with antibodies to COX-2 exhibited a single band (about 72 kDa) in protein extracts from microdissected CA1 sectors or dentate gyrus regions of the hippocampus of control and experimental rats. COX-2 expression in the CA1 regions obtained from rats subjected to 3 or 10 min of ischemia
was increased as early as 4 h and reached a maximum level in 1 day, and decreased thereafter (Fig. 3A and C). However, 10 min of ischemia resulted in a more pronounced expression of COX-2 in the CA1 region (1 day, six-fold of the control level) compared with rats subjected to 3 min of ischemia (1 day, 2.8-fold). When COX-2 protein in the dentate gyrus region in animals subjected to 3 or 10 min of ischemia was determined by immunoblotting, a similar temporal pattern in both groups was observed. After a 3-min ischemia, COX-2 protein in the dentate gyrus regions was significantly increased by 4 h, and reached a 3.5-fold level of induction at 12 h. The level gradually declined thereafter, but this enhanced expression was maintained up to day 14, the latest time point examined (Fig. 3B). In animals subjected to 10 min of ischemia, the temporal pattern for COX2 was similar to that in rats subjected to 3 min of ischemia, although COX-2 expression returned to the basal level by day 7 (Fig. 3D). This study is the first to provide a detailed characterization of the time course and cellular localization of COX-2 in a rat model of ischemic tolerance. Distinct expression patterns of COX-2 were associated with two different ischemic insults: delayed neuronal death and ischemic tolerance in the hippocampal CA1 region.
Fig. 3. Time course of COX-2 protein expression in the microdissected CA1 sectors (A and C) and dentate gyrus regions (B and D) of the hippocampus following 3 min (A and B) or 10 min (C and D) of ischemia. Protein extracts were prepared from rats subjected to sham operation and experimental rats at 4 and 12 h, and 1, 3, 7, and 14 days after ischemia. Lower panels in (A–D): blots were quantified by densitometric analysis. Relative optical densities of the protein bands were obtained from three independent experiments each performed in triplicate and data were expressed as the mean ± S.E.M. * p < 0.01; ** p < 0.005 compared with sham-operated controls.
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
Lethal ischemia in the present study induced significantly increased immunoreactivity for COX-2 in CA1 and hilar neurons in the rat hippocampus. These immunohistochemical data were consistent with previous data, suggesting that COX-2 expression in the vulnerable CA1 and hilar neurons is involved with neuronal death [11,12]. Moreover, our data provide the first in vivo evidence that the increased immunoreactivity in tolerant CA1 pyramidal neurons after the short cerebral-ischemic preconditioning, as well as after the second ischemia with ischemic preconditioning was more transient and significantly weaker than those after lethal ischemia; these results are supported by previous data showing down-regulation of COX-2 after preconditioning in the gerbil hippocampus, as assessed by RT-PCR [3]. The immunohistochemical data were consistent with those obtained by immunoblot analysis, indicating that the upregulation of COX-2 in the microdissected CA1 region was more pronounced in rats subjected to 10 min of ischemia than in rats with 3 min of ischemia. The major finding of this study is that upregulation of COX2 was also observed in dentate granule cells and in CA2 and CA3 neurons following a 3-min period of ischemia, with such results even observed in the same neuronal populations in the ischemic hippocampus. Overall, COX-2 immunoreactivities in these neurons in the tolerant and ischemic hippocampi, were upregulated within 4 h, peaked at 1–3 days and returned to the basal level in 7 days. Upregulation of COX-2 in dentate granule cells and in CA2 and CA3 neurons, which are relatively resistant to ischemic insult [18], is not necessarily followed by cell death, suggesting that the increase of COX-2 is not predictive of cell death in general. Therefore, COX-2 expression in specific neuronal subpopulations of the hippocampus is apparently not involved with neuronal death, thereby indicating a differential sensitivity of neuronal subpopulations to COX-2 upregulation. In regard to the regulation of COX-2 after preconditioning ischemia, Horiguchi et al. [7] suggested that immediate upregulation of COX-2 after ischemic tolerance induction may be involved in the development of ischemic tolerance, although nitric oxide might play a key role in the establishment of preconditioning via an inhibitory effect on subsequent cellular responses involving COX-2. Recently, COX-2 has been reported as an important regulator of enhanced proliferation of neural progenitor cells after ischemic insults [19,21], suggesting that upregulation of COX-2 in dentate granule cells might be involved in hippocampal neurogenesis in the dentate gyrus. However, the functional significance of COX-2 upregulation in these ischemia-resistant neurons requires further investigation. In summary, our results demonstrated that COX-2 expression was substantially and significantly upregulated in vulnerable CA1 and hilar neurons after lethal ischemia, and that the short cerebral-ischemic preconditioning attenuated the increased COX-2 immunoreactivity in these neurons. Upregulation of COX-2 was also induced in resistant CA3 and dentate granule cells in the ischemic and tolerant hippocampi. These results suggest that hippocampal neuronal subpopulations have differential sensitivity to COX-2 upregulation.
235
Acknowledgements This research was supported by a grant (M103KV010019 04K2201 01930) from Brain Research Center of the 21st Century Frontier Research Program funded by the Ministry of Science and Technology of Republic of Korea. We gratefully acknowledge the technical assistance of Hee-Duk Rho and Byung-Ouk Hong. References [1] D. Alcindor, J.G. Krolikowski, P.S. Pagel, D.C. Warltier, J.R. Kersten, Cyclooxygenase-2 mediates ischemic, anesthetic, and pharmacologic preconditioning in vivo, Anesthesiology 100 (2004) 547–554. [2] J.S. Choi, S.Y. Kim, J.H. Cha, Y.S. Choi, K.W. Sung, S.T. Oh, O.N. Kim, J.W. Chung, M.H. Chun, S.B. Lee, M.Y. Lee, Upregulation of gp130 and STAT3 activation in the rat hippocampus following transient forebrain ischemia, Glia 41 (2003) 237–246. [3] V. Colangelo, W.C. Gordon, P.K. Mukherjee, P. Trivedi, P. Ottino, Downregulation of COX-2 and JNK expression after induction of ischemic tolerance in the gerbil brain, Brain Res. 1016 (2004) 195–200. [4] S. Dore, T. Otsuka, T. Mito, N. Sugo, T. Hand, L. Wu, P.D. Hurn, R.J. Traystman, K. Andreasson, Neuronal overexpression of cyclooxygenase2 increases cerebral infarction, Ann. Neurol. 54 (2003) 155–162. [5] S. Govoni, E. Masoero, L. Favalli, A. Rozza, R. Scelsi, S. Viappiani, C. Buccellati, A. Sala, G. Folco, The cycloxygenase-2 inhibitor SC58236 is neuroprotective in an in vivo model of focal ischemia in the rat, Neurosci. Lett. 303 (2001) 91–94. [6] O. Hayaishi, Molecular mechanisms of sleep-wake regulation: roles of prostaglandins D2 and E2, FASEB J. 5 (1991) 2575–2581. [7] T. Horiguchi, J.A. Snipes, B. Kis, K. Shimizu, D.W. Busija, The role of nitric oxide in the development of cortical spreading depression-induced tolerance to transient focal cerebral ischemia in rats, Brain Res. 1039 (2005) 84–89. [8] C. Iadecola, C. Forster, S. Nogawa, H.B. Clark, M.E. Ross, Cyclooxygenase-2 immunoreactivity in the human brain following cerebral ischemia, Acta Neuropathol. 98 (1999) 9–14. [9] W.E. Kaufmann, D.I. Andreasson, P.C. Isakson, P.F. Worley, Cyclooxygenases and the central nervous system, Prostaglandins 54 (1997) 601–624. [10] K. Kitagawa, M. Matsumoto, M. Tagaya, R. Hata, H. Ueda, M. Niinobe, N. Handa, R. Fukunaga, K. Kimura, K. Mikoshiba, T. Kamada, ‘Ischemic tolerance’ phenomenon found in the brain, Brain Res. 528 (1990) 21–24. [11] J. Koistinaho, S. Koponen, P.H. Chan, Expression of cyclooxygenase2 mRNA after global ischemia is regulated by AMPA receptors and glucocorticoids, Stroke 30 (1999) 1900–1906. [12] Y. Matsuoka, M. Okazaki, H. Zhao, S. Asai, K. Ishikawa, Y. Kitamura, Phosphorylation of c-Jun and its localization with heme oxygenase-1 and cyclooxygenase-2 in CA1 pyramidal neurons after transient forebrain ischemia, J. Cereb. Blood Flow Metab. 19 (1999) 1247–1255. [13] S. Miettinen, F.R. Fusco, J. Yrjanheikki, R. Keinanen, T. Hirvonen, R. Roivainen, M. Narhi, T. Hokfelt, J. Koistinaho, Spreading depression and focal brain ischemia induce cyclooxygenase-2 in cortical neurons through N-methyl-d-aspartic acid-receptors and phospholipase A2, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 6500–6505. [14] M. Nakayama, K. Uchimura, R.L. Zhu, T. Nagayama, M.E. Rose, R.A. Stetler, P.C. Isakson, J. Chen, S.H. Graham, Cyclooxygenase-2 inhibition prevents delayed death of CA1 hippocampal neurons following global ischemia, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 10954–11099. [15] S. Nogawa, F. Zhang, M.E. Ross, C. Iadecola, Cyclo-oxygenase-2 gene expression in neurons contributes to ischemic brain damage, J. Neurosci. 17 (1997) 2746–2755. [16] M.A. Perez-Pinzon, G.P. Xu, P.L. Mumford, W.D. Dietrich, M. Rosenthal, T.J. Sick, Rapid ischemic preconditioning protects rats from cerebral anoxia/ischemia, Adv. Exp. Med. Biol. 428 (1997) 155–161.
236
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
[17] W.A. Pulsinelli, J.B. Brierley, A new model of bilateral hemispheric ischemia in the unanesthetized rat, Stroke 10 (1979) 267–272. [18] W.A. Pulsinelli, D.E. Levy, T.E. Duffy, Regional cerebral blood flow and glucose metabolism following transient forebrain ischemia, Ann. Neurol. 11 (1982) 499–502. [19] T. Sasaki, K. Kitagawa, S. Sugiura, E. Omura-Matsuoka, S. Tanaka, Y. Yagita, H. Okano, M. Matsumoto, M. Hori, Implication of cyclooxygenase-2 on enhanced proliferation of neural progenitor cells
in the adult mouse hippocampus after ischemia, J. Neurosci. Res. 72 (2003) 461–471. [20] W.L. Smith, R.M. Garavito, D.L. DeWitt, Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2, J. Biol. Chem. 271 (1996) 33157–33160. [21] K. Uchida, K. Kumihashi, S. Kurosawa, T. Kobayashi, K. Itoi, T. Machida, Stimulatory effects of prostaglandin E2 on neurogenesis in the dentate gyrus of the adult rat, Zoolog. Sci. 19 (2002) 1211–1216.
Differential regulation of cyclooxygenase-2 in the rat hippocampus after cerebral ischemia and ischemic tolerance Jeong-Sun Choi, Ha-Young Kim, Myung-Hoon Chun, Jin-Woong Chung, Mun-Yong Lee ∗ Department of Anatomy, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Socho-gu, Seoul 137-701, Republic of Korea Received 25 July 2005; received in revised form 14 September 2005; accepted 28 September 2005
Abstract We investigated the temporal changes and cellular localization of cyclooxygenase-2 (COX-2) in the rat hippocampus during the induction of acquired ischemic tolerance by sublethal ischemia, and compared these changes with those occurring following transient forebrain ischemia. Adult male Sprague Dawley rats were subjected to either 10 min of lethal global ischemia with or without 3 min of sublethal ischemic preconditioning, or 3 min of ischemia only. A short (3 min) cerebral ischemia as well as lethal ischemia with preconditioning substantially and significantly upregulated COX-2 expression in dentate granule cells, as confirmed by immunoblot analysis. This became evident by 4 h, peaked at 1–3 days, and returned to the basal level around 7 days. COX-2 expression was also increased in CA2 and CA3 neurons, although with weaker staining intensity, but in CA1 neurons very weak immunoreactivity was transiently observed. In the ischemic hippocampus, however, in agreement with previous reports, COX-2 expression was induced strongly in vulnerable CA1 and hilar neurons as well as in resistant CA3 and dentate granule cells. These data demonstrated that COX-2 expression is upregulated in neuronal subpopulations destined to survive, i.e., in CA3 and dentate granule cells after ischemia and ischemia-tolerance induction, as well as in ischemia-vulnerable neurons, i.e., in CA1 neurons after lethal ischemia, suggesting that hippocampal neuronal subpopulations have differential sensitivity to COX-2 upregulation. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: COX-2; Hippocampus; Ischemic preconditioning; Dentate granule cells
Cyclooxygenase (COX), a rate-limiting enzyme in the synthesis of prostaglandins (PG) from arachidonic acid, produces PGH2, which in subsequent steps gives rise to PGs with various physiological functions [6,9,20]. COX-2, the inducible form, is thought to be involved in mechanisms of ischemic brain injury. Cerebral ischemia upregulates COX-2 in neurons, glial cells and infiltrating leukocytes in injured brain [8,13,15]. In addition, pharmacological inhibition of COX-2 and genetic deletion of COX-2 reduce infarct size, and neuronal overexpression of COX-2 increases cerebral infarction [4,5,14,15]. These observations suggest that COX-2 plays a deleterious role in ischemic injury. Recently, Horiguchi et al. [7] reported that ischemic tolerance by cortical spreading depression induced immediate upregulation of COX-2. In addition, COX-2 is a critical mediator of myocardial protection by ischemic preconditioning [1]. Although upregulation of COX-2 in the ischemic hippocampus
∗
Corresponding author. Tel.: +82 2 590 1108; fax: +82 2 536 3110. E-mail address: [email protected] (M.-Y. Lee).
0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.09.074
is attenuated by induction of ischemic tolerance using real-time polymerase chain reaction (RT-PCR) [3], little information is known about the selective expression and regulation of COX-2 after a preconditioning ischemic stimulus. A brief episode of sublethal ischemia is known to provide resistance against damage induced by a subsequent lethal ischemia [10,16]. This study was therefore performed to investigate the temporal changes and cellular localization of COX-2 in the rat hippocampus during the induction of acquired ischemic tolerance by sublethal ischemia. We also compared these changes with those occurring following transient forebrain ischemia. Transient forebrain ischemia was induced by four-vessel occlusion and reperfusion, as previously described [17] with minor modifications [2]. Briefly, the vertebral arteries were electrocauterized and cut completely to stop circulation in these vessels. After 24 h, both common carotid arteries were occluded for 3 min (sublethal ischemia) or 10 min (lethal ischemia) using miniature aneurismal clips. In some experiments, a 3-min period of occlusion (ischemic preconditioning) or sham operation was followed by 3 days of reperfusion, after which 10 min of occlusion was induced again. Only animals with complete EEG
232
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
flattening upon vascular occlusion were classified as ischemic and used in the study. Rectal temperature was maintained at 37.5 ± 0.3 ◦ C using a heating lamp during and after ischemia. Sham-operated rats with cauterized vertebral arteries and ligatures placed around the carotid arteries were used as controls. No animal convulsed or died following reperfusion or sham operation. Animals were sacrificed 4 and 12 h, and 1, 3, 7, and 14 days after reperfusion. For each time point, three rats were used for immunoblot analysis and seven for immunohistochemistry. Sham-operated animals were treated using the same schedule as the ischemic-reperfused animals. At each time point after reperfusion, animals were deeply anesthetized with 16.9% urethane (10 mL/kg) and killed by transcardial perfusion with a fixative containing 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) or by decapitation followed by dissection of the hippocampi, which were quickly frozen in liquid nitrogen. All experimental procedures performed on the animals were conducted with the approval of the Catholic Ethics Committee of the Catholic University of Korea, and were in accordance with the National Institute of Health’s Guide for the Care and Use
of Laboratory Animals (NIH Publication No. 80-23, revised 1996). For immunoblot analysis, coronal slices (400 m thick) were cut at the level of the septal hippocampus with a McIlwain tissue chopper. The dorsal hippocampal CA1 sectors and dentate gyrus regions were selectively dissected from the slices under microscopic observation, and were homogenized in ice-cold RIPA buffer (50 mM Tris buffer, pH 8.0; 150 mM NaCl; 1% NP-40; 0.5% deoxycholate; and 0.1% SDS). Equal amounts (20 g) of total protein from the CA1 sector or dentate gyrus of the hippocampi from sham-operated rats and rats that had undergone 3 or 10 min of ischemia were separated in 8% SDS–polyacrylamide gel electrophoresis (PAGE) and blotted onto cellulose membranes. Immunostaining of the blots was performed using a goat polyclonal antibody against COX-2 (Santa Cruz Biotechnology, Santa Cruz, CA, USA; diluted 1:1000). The immunoreactive bands were detected by enhanced chemiluminescence (ECL; Amersham, Arlington Heights, IL, USA). Densitometric analysis was performed using the Eagle Eye TMII Still Video System (Stratagene, La Jolla, CA, USA). Three
Fig. 1. Comparison of COX-2 expression in the rat hippocampus between ischemic tolerance (B, C, E and G) and cerebral ischemia (D, F and H). (A) In control sections of the hippocampus, weak COX-2 immunoreactivity was localized to neurons of the pyramidal cell and granule cell layers. DG, the dentate gyrus. (B) Four hours after 3 min of ischemia, COX-2 immunoreactivity increased significantly in the granule cells and slightly in CA2 and CA3 neurons. (C) One day after 3 min of ischemia, increased immunoreactivity in the granule cells was more evident and weakly labeled profiles had appeared in the pyramidal cell layers along CA1–CA3. (D) One day after 10 min of ischemia, COX-2 immunoreactivity had increased preferentially in the CA1 and dentate hilar region and also in granule cells and CA2 and CA3 neurons. (E–H) Higher magnification views of the boxed areas in (C) and (D), respectively. GL, the granule cell layer; pcl, the pyramidal cell layer; PL, the dentate hilar region; sr, the stratum radiatum. Scale bar = 300 m for (A–D); 50 m for (E–F); 100 m for (G–H).
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
animals were used for immunoblotting at each time point, and relative optical densities of the protein bands were obtained from three independent experiments each performed in triplicate. The data were expressed as mean ± S.E.M. Statistical significance was analyzed by analysis of variance (ANOVA) followed by Dunnett’s t-test; p < 0.01 was regarded as significant. For COX-2 immunohistochemistry, free-floating, 25-mthick sections were processed. After blocking with 10% normal donkey serum for 1 h, the sections were incubated overnight at 4 ◦ C with a goat polyclonal antibody against COX-2 (Santa Cruz Biotechnology; diluted 1:100). Primary antibody binding was visualized using peroxidase-labeled donkey anti-goat antibody (Jackson ImmunoResearch, West Grove, PA, USA; diluted 1:200), and 0.05% 3,3 -diaminobenzidine tetrahydrochloride and 0.01% H2 O2 as the substrate. The specificity of COX-2 immunoreactivities was confirmed by the absence of immunohistochemical reaction in sections from which the primary antibody was omitted or substituted with non-specific rabbit IgG. Histological staining with Cresyl violet was performed on sections adjacent to those processed for immunohistochemistry.
233
The distribution and cellular localization of COX-2 immunoreactivity in the hippocampus following 3 or 10 min of ischemia were examined using immunohistochemistry. A low level of basal COX-2 immunoreactivity occurred in neurons of the pyramidal cell and granule cell layers in the hippocampus of control rats at all time points (Fig. 1A). By 4 h after 3 min of ischemia, COX-2 immunoreactivity was significantly increased in dentate granule cells but only slightly increased in CA2 and CA3 neurons (Fig. 1B). By 1 day after 3 min of ischemia, COX2 immunoreactivity was progressively enhanced in granule cells and also more evident in CA2 and CA3 neurons (Fig. 1C and G). In addition, more weakly labeled profiles had appeared in the pyramidal cell layers in the CA1 region (Fig. 1E). In the ischemic hippocampus after 10 min of ischemia, COX-2 expression was upregulated in all hippocampal neurons after 1 day of reperfusion (Fig. 1D). The signals in CA1 neurons and the dentate hilar neurons were stronger than those of the dentate granule cells (Fig. 1F and H). Three days after 3 min of ischemia, increased immunoreactivity was still observed in dentate granule cells and in CA2 and
Fig. 2. Comparison of COX-2 expression in the rat hippocampus between ischemic tolerance (A, C, E and G) and cerebral ischemia (B, D, F and H). (A) Three days after 3 min of ischemia, increased COX-2 immunoreactivity was observed in granule cells and in CA2 and CA3 neurons. (B) Three days after 10 min of ischemia, intense immunoreactivity was observed in granule cells and CA3 neurons, and also in scattered surviving neurons in the CA1 region. (C) Seven days after the 3 min of ischemia, COX-2 immunoreactivity had returned toward control levels. (D) Seven days after the 10 min ischemic insult, neuronal COX-2 expression had returned to the control level, but immunoreactivity was observed in small cells showing the morphology of astrocytes, which were sparsely distributed in the strata radiatum and oriens of the CA1 region. (E–H) Higher magnification views of the boxed areas in (A–D). pcl, the pyramidal cell layer; sr, the stratum radiatum. Scale bar = 300 m for (A–D); 50 m for (E–H).
234
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
CA3 neurons, whereas no significant labeling was observed in preserved CA1 neurons (Fig. 2A and E). Three days after 10 min of ischemia, intense COX-2 immunoreactivity was observed in CA3 and granule cells, and in a few scattered surviving neurons in the CA1 region where most neuronal signals had disappeared (Fig. 2B and F). Preconditioned hippocampus that had been subjected to a subsequent 10 min of ischemia exhibited a labeling pattern similar to that seen in the hippocampus following 3 min of ischemia (data not shown). COX-2 immunoreactivity had returned toward control levels in 7 days (Fig. 2C and G) and 14 days (data not shown) after 3 min of ischemia. Seven days after 10 min of ischemia, the expression of COX-2 had returned to the control level, with a few labeled cells that had the morphology of astrocytes observed in the strata radiatum and oriens of the CA1 region (Fig. 2D and H), and the labeling pattern remained unchanged for 14 days (data not shown). The results of the immunoblot analysis after 3 or 10 min of ischemia were consistent with the immunohistochemical data. Immunoblot analyses with antibodies to COX-2 exhibited a single band (about 72 kDa) in protein extracts from microdissected CA1 sectors or dentate gyrus regions of the hippocampus of control and experimental rats. COX-2 expression in the CA1 regions obtained from rats subjected to 3 or 10 min of ischemia
was increased as early as 4 h and reached a maximum level in 1 day, and decreased thereafter (Fig. 3A and C). However, 10 min of ischemia resulted in a more pronounced expression of COX-2 in the CA1 region (1 day, six-fold of the control level) compared with rats subjected to 3 min of ischemia (1 day, 2.8-fold). When COX-2 protein in the dentate gyrus region in animals subjected to 3 or 10 min of ischemia was determined by immunoblotting, a similar temporal pattern in both groups was observed. After a 3-min ischemia, COX-2 protein in the dentate gyrus regions was significantly increased by 4 h, and reached a 3.5-fold level of induction at 12 h. The level gradually declined thereafter, but this enhanced expression was maintained up to day 14, the latest time point examined (Fig. 3B). In animals subjected to 10 min of ischemia, the temporal pattern for COX2 was similar to that in rats subjected to 3 min of ischemia, although COX-2 expression returned to the basal level by day 7 (Fig. 3D). This study is the first to provide a detailed characterization of the time course and cellular localization of COX-2 in a rat model of ischemic tolerance. Distinct expression patterns of COX-2 were associated with two different ischemic insults: delayed neuronal death and ischemic tolerance in the hippocampal CA1 region.
Fig. 3. Time course of COX-2 protein expression in the microdissected CA1 sectors (A and C) and dentate gyrus regions (B and D) of the hippocampus following 3 min (A and B) or 10 min (C and D) of ischemia. Protein extracts were prepared from rats subjected to sham operation and experimental rats at 4 and 12 h, and 1, 3, 7, and 14 days after ischemia. Lower panels in (A–D): blots were quantified by densitometric analysis. Relative optical densities of the protein bands were obtained from three independent experiments each performed in triplicate and data were expressed as the mean ± S.E.M. * p < 0.01; ** p < 0.005 compared with sham-operated controls.
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
Lethal ischemia in the present study induced significantly increased immunoreactivity for COX-2 in CA1 and hilar neurons in the rat hippocampus. These immunohistochemical data were consistent with previous data, suggesting that COX-2 expression in the vulnerable CA1 and hilar neurons is involved with neuronal death [11,12]. Moreover, our data provide the first in vivo evidence that the increased immunoreactivity in tolerant CA1 pyramidal neurons after the short cerebral-ischemic preconditioning, as well as after the second ischemia with ischemic preconditioning was more transient and significantly weaker than those after lethal ischemia; these results are supported by previous data showing down-regulation of COX-2 after preconditioning in the gerbil hippocampus, as assessed by RT-PCR [3]. The immunohistochemical data were consistent with those obtained by immunoblot analysis, indicating that the upregulation of COX-2 in the microdissected CA1 region was more pronounced in rats subjected to 10 min of ischemia than in rats with 3 min of ischemia. The major finding of this study is that upregulation of COX2 was also observed in dentate granule cells and in CA2 and CA3 neurons following a 3-min period of ischemia, with such results even observed in the same neuronal populations in the ischemic hippocampus. Overall, COX-2 immunoreactivities in these neurons in the tolerant and ischemic hippocampi, were upregulated within 4 h, peaked at 1–3 days and returned to the basal level in 7 days. Upregulation of COX-2 in dentate granule cells and in CA2 and CA3 neurons, which are relatively resistant to ischemic insult [18], is not necessarily followed by cell death, suggesting that the increase of COX-2 is not predictive of cell death in general. Therefore, COX-2 expression in specific neuronal subpopulations of the hippocampus is apparently not involved with neuronal death, thereby indicating a differential sensitivity of neuronal subpopulations to COX-2 upregulation. In regard to the regulation of COX-2 after preconditioning ischemia, Horiguchi et al. [7] suggested that immediate upregulation of COX-2 after ischemic tolerance induction may be involved in the development of ischemic tolerance, although nitric oxide might play a key role in the establishment of preconditioning via an inhibitory effect on subsequent cellular responses involving COX-2. Recently, COX-2 has been reported as an important regulator of enhanced proliferation of neural progenitor cells after ischemic insults [19,21], suggesting that upregulation of COX-2 in dentate granule cells might be involved in hippocampal neurogenesis in the dentate gyrus. However, the functional significance of COX-2 upregulation in these ischemia-resistant neurons requires further investigation. In summary, our results demonstrated that COX-2 expression was substantially and significantly upregulated in vulnerable CA1 and hilar neurons after lethal ischemia, and that the short cerebral-ischemic preconditioning attenuated the increased COX-2 immunoreactivity in these neurons. Upregulation of COX-2 was also induced in resistant CA3 and dentate granule cells in the ischemic and tolerant hippocampi. These results suggest that hippocampal neuronal subpopulations have differential sensitivity to COX-2 upregulation.
235
Acknowledgements This research was supported by a grant (M103KV010019 04K2201 01930) from Brain Research Center of the 21st Century Frontier Research Program funded by the Ministry of Science and Technology of Republic of Korea. We gratefully acknowledge the technical assistance of Hee-Duk Rho and Byung-Ouk Hong. References [1] D. Alcindor, J.G. Krolikowski, P.S. Pagel, D.C. Warltier, J.R. Kersten, Cyclooxygenase-2 mediates ischemic, anesthetic, and pharmacologic preconditioning in vivo, Anesthesiology 100 (2004) 547–554. [2] J.S. Choi, S.Y. Kim, J.H. Cha, Y.S. Choi, K.W. Sung, S.T. Oh, O.N. Kim, J.W. Chung, M.H. Chun, S.B. Lee, M.Y. Lee, Upregulation of gp130 and STAT3 activation in the rat hippocampus following transient forebrain ischemia, Glia 41 (2003) 237–246. [3] V. Colangelo, W.C. Gordon, P.K. Mukherjee, P. Trivedi, P. Ottino, Downregulation of COX-2 and JNK expression after induction of ischemic tolerance in the gerbil brain, Brain Res. 1016 (2004) 195–200. [4] S. Dore, T. Otsuka, T. Mito, N. Sugo, T. Hand, L. Wu, P.D. Hurn, R.J. Traystman, K. Andreasson, Neuronal overexpression of cyclooxygenase2 increases cerebral infarction, Ann. Neurol. 54 (2003) 155–162. [5] S. Govoni, E. Masoero, L. Favalli, A. Rozza, R. Scelsi, S. Viappiani, C. Buccellati, A. Sala, G. Folco, The cycloxygenase-2 inhibitor SC58236 is neuroprotective in an in vivo model of focal ischemia in the rat, Neurosci. Lett. 303 (2001) 91–94. [6] O. Hayaishi, Molecular mechanisms of sleep-wake regulation: roles of prostaglandins D2 and E2, FASEB J. 5 (1991) 2575–2581. [7] T. Horiguchi, J.A. Snipes, B. Kis, K. Shimizu, D.W. Busija, The role of nitric oxide in the development of cortical spreading depression-induced tolerance to transient focal cerebral ischemia in rats, Brain Res. 1039 (2005) 84–89. [8] C. Iadecola, C. Forster, S. Nogawa, H.B. Clark, M.E. Ross, Cyclooxygenase-2 immunoreactivity in the human brain following cerebral ischemia, Acta Neuropathol. 98 (1999) 9–14. [9] W.E. Kaufmann, D.I. Andreasson, P.C. Isakson, P.F. Worley, Cyclooxygenases and the central nervous system, Prostaglandins 54 (1997) 601–624. [10] K. Kitagawa, M. Matsumoto, M. Tagaya, R. Hata, H. Ueda, M. Niinobe, N. Handa, R. Fukunaga, K. Kimura, K. Mikoshiba, T. Kamada, ‘Ischemic tolerance’ phenomenon found in the brain, Brain Res. 528 (1990) 21–24. [11] J. Koistinaho, S. Koponen, P.H. Chan, Expression of cyclooxygenase2 mRNA after global ischemia is regulated by AMPA receptors and glucocorticoids, Stroke 30 (1999) 1900–1906. [12] Y. Matsuoka, M. Okazaki, H. Zhao, S. Asai, K. Ishikawa, Y. Kitamura, Phosphorylation of c-Jun and its localization with heme oxygenase-1 and cyclooxygenase-2 in CA1 pyramidal neurons after transient forebrain ischemia, J. Cereb. Blood Flow Metab. 19 (1999) 1247–1255. [13] S. Miettinen, F.R. Fusco, J. Yrjanheikki, R. Keinanen, T. Hirvonen, R. Roivainen, M. Narhi, T. Hokfelt, J. Koistinaho, Spreading depression and focal brain ischemia induce cyclooxygenase-2 in cortical neurons through N-methyl-d-aspartic acid-receptors and phospholipase A2, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 6500–6505. [14] M. Nakayama, K. Uchimura, R.L. Zhu, T. Nagayama, M.E. Rose, R.A. Stetler, P.C. Isakson, J. Chen, S.H. Graham, Cyclooxygenase-2 inhibition prevents delayed death of CA1 hippocampal neurons following global ischemia, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 10954–11099. [15] S. Nogawa, F. Zhang, M.E. Ross, C. Iadecola, Cyclo-oxygenase-2 gene expression in neurons contributes to ischemic brain damage, J. Neurosci. 17 (1997) 2746–2755. [16] M.A. Perez-Pinzon, G.P. Xu, P.L. Mumford, W.D. Dietrich, M. Rosenthal, T.J. Sick, Rapid ischemic preconditioning protects rats from cerebral anoxia/ischemia, Adv. Exp. Med. Biol. 428 (1997) 155–161.
236
J.-S. Choi et al. / Neuroscience Letters 393 (2006) 231–236
[17] W.A. Pulsinelli, J.B. Brierley, A new model of bilateral hemispheric ischemia in the unanesthetized rat, Stroke 10 (1979) 267–272. [18] W.A. Pulsinelli, D.E. Levy, T.E. Duffy, Regional cerebral blood flow and glucose metabolism following transient forebrain ischemia, Ann. Neurol. 11 (1982) 499–502. [19] T. Sasaki, K. Kitagawa, S. Sugiura, E. Omura-Matsuoka, S. Tanaka, Y. Yagita, H. Okano, M. Matsumoto, M. Hori, Implication of cyclooxygenase-2 on enhanced proliferation of neural progenitor cells
in the adult mouse hippocampus after ischemia, J. Neurosci. Res. 72 (2003) 461–471. [20] W.L. Smith, R.M. Garavito, D.L. DeWitt, Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2, J. Biol. Chem. 271 (1996) 33157–33160. [21] K. Uchida, K. Kumihashi, S. Kurosawa, T. Kobayashi, K. Itoi, T. Machida, Stimulatory effects of prostaglandin E2 on neurogenesis in the dentate gyrus of the adult rat, Zoolog. Sci. 19 (2002) 1211–1216.