Brain Research, 587 (1992) 195-202
195
© 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 17953
Distributions of heat shock protein (HSP) 70 and heat shock cognate protein (HSC) 70 mRNAs after transient focal ischemia in rat brain J. Kawagoe, K. Abe, S. Sato, I. Nagano, S. N a k a m u r a and K. K o g u r e Department of Neurololo, Institute of Brain Diseases, Tohoku University School of Medicine, Sendal (Japan) (Accepted 10 March 1992)
Key words: Heat shock protein; lteat shock cognate protein; Cerebral focal ischemia; in situ hybridization; Rat
The distribution of heat shock protein (HSP) 70 and heat shock cognate protein (HSC) 70 mRNA after 30 min of middle cerebral artery (MCA) occlusion was investigated in rat brain by in situ hybridization using cloned eDNA probes s~.lective for the mRNAs. While HSP70 mRNA was hardly present at caudate and dorsal hippocampal levels of the sham brain this mRNA was greatly induced in cells of the MCA territory ! h after reperfusion. Although the maximum amount of induced HSP70 mRNA in the caudate was much smaller than that in the cortex the maximum induction in the caudate (3 h) preceded that in the cortex (8 h). in contrast to the case of HSP70 mRNA, HSC70 mRNA was present in most cells of the sham brain, and was especially dense in hippocampal CA3 cells. Further induction of HSC70 mRNA was observed after reperfusion in the same cell populations, as in the case of HSP70 mRNA. HSC70 mRNA levels were significantly reduced in the caudate at 8 h when small amounts of HSPT0 mRNA were still elevated. In the ipsilateral granule cells of the dentate gyrus and hippocampal CA3 cells a slight but significant induction of HSC70 mRNA was observed from I h to I day, while obvious induction of HSPT0 mRNA never occurred. All the induced signals of HSPT0 and HSC70 mRNA were diminished or returned to the sham level by 7 days, except for HSC70 mRNA in the caudate. These results are the first observations of the distribution of HSP70 and HSC70 mRNA after transient focal ischemia of rat brain. The results indicate that the temporal and quantitative differences in the maximum induction of HSP70 and HSC70 mRNA may relate to the different susceptibilities between cells of the caudate and cortex. Different roles under normal conditions and a general cooperative role in the recovery process from ischemic injury between HSP70 and itSC7(l are suggested. The finding of selective induction of HSC70 mRNA in hippocampal cells may indicate that a stress response, which is not accompanied by tlSP70 mRNA induction, occurs in the brain after transient ischemia.
INTRODUCTION Stressful conditions of brain, such as tissue injury s, heat shock t~, administration of drugs t.~,t,J and ischemia 2,.~,t.~,t4,t~,a,,aa,2s induce 'stress responses', including the synthesis of the 70 kDa heat-shock protein (HSP70). HSP70 has been suggested to play a protective role under stressful conditions of cells 3.~,.t~,,tg.23.as. Recent studies suggested regional differences in immunoreactivity for HSP70 protein among the hippocampal cell populations corresponded to different vulnerabilities after transient global ischemia in gerbil t¢,,28. These results indicate an important role for HSP70 in protecting cells against injury and/or promoting the recovery of injured cells. The degree of cell damage also differs among cell populations of the ischemic territory after a middle cerebral artery (MCA) occlusion in rat brain 1,20.
Although the distribution of induced HSP70 mRNA has been reported in the global ischemia model of gerbil 22 there has been no similar report in a focal cerebral ischemia model in rat, Thus it is of interest to investigate the change in the spatial distribution of HSP70 mRNA after focal cerebral ischemia in rat for a proposal of a mechanism for the regional differences in cell damage. Heat shock cognate protein (HSC) 70 is a member of the HSP70 family, and is constitutively expressed during normal cell development and differentiation I I The induction of HSC70 mRNA has been investigated under normal and stressful conditions, including ischemia in gerbil and rat brain, by Northern blot analysis 3.tt~.2t. Abe et al. a observed a regional difference in the induction of HSC70 mRNA after transient ischemia, and proposed a possible protective role for
Correspondence: J.-i. Kawagoe, Department of Neurology, Institute of Brain Diseases, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendal 980, Japan. Fax: (81) (22) 272-5818.
196
HSC70 protein in cell damage after ischemia. However, the distribution of HSC70 mRNA in subregions of normal and post-ischemic rat brain has not yet been investigated. We recently isolated eDNA clones selective for HSP70 and HSC70 mRNAs from gerbil brain -'~'.Using these eDNA clones as probes we investigated the spatial distribution of HSP70 and HSC70 mRNA by in situ hybridization in rat brain after 30 min of focal ischemia. MATERIALS AND METHODS
at a concentration of 0.1 n g / # l in prehybridization solution supplemented with 10% dextran sulfate and I0 mM dithiothreitol. The eDNA inserts were radiolabeled with [a -3sS]deoxycytidine triphosphate (I,000 Ci/mmol, Amersham) by random-primer labeling io using a kit (Boehringer Mannheim Yamanouchi Co., Tokyo, Japan), resulting in specific activities of 5 - 6 x 10~ dpm//zg. The control probe was a similarly labeled plasmid vector (pHSG396, Takara Shuzo Co., Kyoto, Japan) fragment (1.1 kb) digested by Hinfl. The sections were washed at room temperature for 2 h in 2 x SSC and for ! h in ! x SSC, and were then dehydrated through a graded ethanol series containing 0.3 M ammonium acetate. After visualization of the hybridization by exposure to X-ray film for 24 h at room temperature the sections were dipped in a liquid emulsion (NR-M2, Konica Co., Tokyo, Japan) diluted (1:1) with 0.6 M ammonium acetate, They were then exposed at 4°C for approximately 3 weeks, developed, and counterstained with hematoxylin. Several slides were treated with 100 p,g/ml RNAse A and 10 U / m l RNAse TI (Sigma, St. Louis, MO, USA) in 2 x SSC at 37"C for 2 h prior to prehybridization, and were then hybridized with the probe for HSPT0 or HSCT0 mRNA,
.4nunal mo&'l The animal model used in the present study consisted of a middle cerebral artery (MCA) occlusion using a microembolus of nylon thread i,.'. Male Wistar rats weighing 2511-2811 g were lightly anesthetized by inhalation of a nitrous oxide/oxygen/halothane (6t)c; : 31|¢; : If; ) mixture during `surgical preparation. The right MCA wa`s expo`sed and then the anesthesia was stopped. When the animal began to regain consciousness the origin of the artery was occluded h)r 311 min by insertion of a nylon thread via the common carotid artery. Body temperatare was monitored in :dl animals with a rectal probe, and was maintained at 37°C using a heat pad daring the `surgical preparation and occlusion of the MCA. After 30 rain of hchemia the bh)od flow was restored by removal of the nylon thread, and no further attempt was made to maintain constant body temperalure of ,mimah, The animals were allowed It) recover for I, 3. 8 h, and I, 2 and 7 days (n - 3 for each time point) at ambient temper:ilure (21=23°C), and then were de~:apitated, Sham control animals wer~ treated with cervical ,~urtlery but without the followin~ insertion of nyh)n thread.
Histopathologicai study Ischemia-induced cell damage was estimated at 7 days by light microscopic observation of Cresyl violet-stained sections (5/~m) of brains receiving conventional fixation 4,s.
RESULTS The results obtained in hybridization and histopathological studies were reproducible in all animals employed at each time point.
Hybridization experiments Figure ! shows the results of Northern blot analysis using cloned HSP70 and HSC70 eDNA probes against
i b.t)ndi:atiot) (',~7)('rim('nr~ The insem of eh),ed ~DNAs used a,~ probes in the present study weru originally from th~ eerd)r.l cortex t)l' ~erbil, The ~izes of the inserts `selective I'or tlSP711 and HSC711 mRNA were 1.11(pGA ~)and 1.4 kb (pClD~), respectively ~r, The ~pe¢ifi¢ities of these probes for lISP70 and t!SC7(1 mRNA in rat brain were examined by Northern blot analy,~is. Total RNA was extracted from cerebral cortices of the t)c¢luded MCA territory, electrophoresed, and transferred to a nylon membrane, as in our previous report ~, The probes were radiolabeled with [ , - ~ " P ] dATP 16,01H1 Ci/mmol, Amersham, Tokyo, Japan) by random-primer labeling, and hybridized against Northern Idols 115/~g of total RNA/lane) at 42°C for 20 h in a hybridization solution containing formamide, After the hybridization the filters were wa`shed with 2×SSC (I ×SSC - 151)mM NaCI, 15 mM sodium citrate), I x SSC, and 0,5 x SSC with 11,2~ SDS at n)om temperature t)r t)5°C. The filters were exposed to X-ray film for 21) h at -8(I°C, In situ hybridization was performed by the method of Yoshioka el al. -~) with a slight modification. Briefly, the dissected brains were frozen in powdered dry ice and stored at - 800C, Sections ( Ill/,tin) at the caudate and dorsal hippocampal levels were cut on a cryostat at -18°C, and collected onto slides coated with Histostik (Accurate Chemical and Scientific Corp,, We`stbury, NY), The sections were fixed h)r 11) rain in an ethanol/acetic acid (3:1) mixture, transferred to 11.2 M HCI for 211min, and immersed tilt 211rain at 5II°C in 2 x SSC (pH 7.1)). They were then digested h)r 15 rain at 37°C with Hill ptg/ml of proteinase K (Merk & Co,, Rahway, NJ) in 21) mM Tris-HCI buffer (pH 7.4)with 2 raM CaCI,, and dehydrated through a graded ethanul ,series, Slides were prehybridized for 2 h at room temperature in a solution containing 511% tbrmamide, 81~11mM NaCI, IU mM Tris-HCI (pH 7.5), I mM EDTA, 0,02% polyvinylpyrrolidone, (),U2r~ Ficoll, I),02Ci bovine serum albumin, I(X) p,g/ml sonicated, denatured calf thymus DNA, and 1),5 mg/ml calf liver RNA, Hybridization was done for 21) h at 42°C using ~~S-labeled eDNA probe
HSP 70
HSC 70
S
S 8h 2d
8h 2d
Fig, I, Northern blot analysis of HSPT0 and HSC70 mRNAs in rat cerebral cortices of the ipsilateral MCA territory of sham control (S) and post-ischemic brain at 8 h (8h) and 2 days (2d) after the reperfusion. Note that HSPT0 and HSCT0 probes detected the different mRNA species in the rat brain and did not cross react, Small and large arrowheads show positions of the detected HSPT0 (2,8 and 3,0 kb) and HSC70 (2.4 kb) mRNAs, respectively.
197 total RNA isolated from rat cerebral cortex in the present ischemic model. Figure 1 shows the autoradiogram of the membrane filter finally washed with 0.5 x SSC at 65°C. A final wash with 2 or 1 x SSC at room temperature showed the same results (data not shown). In the sham control brain there was no hybridization of HSP70 mRNA, whereas HSC70 mRNA was identified. At 8 h after repeffusion HSP70 mRNA levels were greatly induced, and a slight increase in the HSC70 mRNA level was also observed. Thus HSP70 and HSC70 probes used in the present study selectively
detected the HSP70 mRNA doublet (2.8 and 3.0 kb) and the HSC70 mRNA species (2.4 kb), respectively, in rat brain. The identification patterns of HSP70 and HSC70 mRNAs with eDNA probes (Fig. l) were similar to that in the previous report of Northern blot analysis using oligonucleotide probes specific for HSPT0 and HSC70 mRNAs ~9. By 2 days the amount of HSPT0 mRNA had decreased, and that of HSC70 mRNA had returned to almost the sham control level. Neither the HSP70 probe nor the HSC70 probe cross-hybridized with any other mRNA species.
S
/ •
I h ~
3h
•
~
~,
, •
•
.
'.,,~_" "
~'.t' .,"
..
8h
(,~, '
~.
ld ~,~
,,
.(.,
2d
d
' f"
b ¸
m Fig. 2. Distribution of HSPT0 mRNA at caudate (left column) and dorsal hippocampal (right column) levels of sham (S) and post.ischemic rat brain. Bar = 2 ram.
198 Distribution of HSP70 mRNA in post-ischemic rat brain at caudate and dorsal hippocampal levels is shown in Fig. 2. While HSP70 mRNA was hardly present in the sham brain the mRNA was induced in cells of the caudate and cerebral cortex even 1 h after reperfusion. The HSP70 mRNA induction was gradually intensified, with peaks at 3 h for cells of the caudate and 8 h for the cortex. The maximum amount of induced HSP70 mRNA in the caudate was smaller than that in the cortex, in the caudate the induced HSPT0 mRNA level
was greatly reduced at 8 h, and the hybridization signal was almost diminished at 1 day. All of the induced signal of HSP70 mRNA disappeared by 7 days after reperfusion. Spatial changes of distribution of HSC70 mRNA in sham and post-ischemic rat brain at caudate and dorsal hippocampai levels are documented in Fig. 3. In contrast to HSP70 mRNA, HSC70 mRNA was present in most cell populations of the sham brain. A relatively high level of HSC70 mRNA was observed in cells of
S
lh
3h
8h d I
ld
2d
7d
m
Fig, 3. Distribution of H$C70 mRNA at caudate (left column) and dorsal hippocampal (right column) levels of sham ($) and post-ischemic rat brain. Bar = 2 ram,
199 the medial habenula and hippocampal CA3. The transient focal ischemia increased the amount of HSC70 mRNA in the same cell populations as in HSPT0 mRNA, with the same peak times. In addition to those cell populations HSC70 mRNA was induced in the ipsilateral granule cells of dentate gyrus and hippocampal CA3 cells from 1 h to I day (Figs. 3 and 4). The magnitude of HSC70 mRNA induction was greater in dentate granule cells than CA3 cells, however, CA1 cells were never affected. At 8 h after reperfusion HSC70 mRNA levels were greatly reduced in the caudate (Fig. 3) while a small amount of HSP70 mRNA was still elevated (Fig. 2). The significant reduction of HSC70 mRNA levels in the caudate continued to 7 days. The induced HSC70 mRNA returned to the sham level by 7 days except for the caudate. Observation of the sections dipped in the liquid emulsion revealed that grains for HSP70 and HSC70 probes were predominantly located in the cell body (data not shown). Neither the sections incubated with plasmid vector probe nor the sections treated with RNAses before incubation with the probe for HSP70 or HSC70 mRNA exhibited any specific hybridization (data not shown).
HSC70
S lh ld Fig. 4. High magnificationphotographsof the spatial change in the distribution of HSCT0mRNA at the dorsal hippocampallevel. The sections of sham (S) and post-ischemicbrain at I h (lh) and 1 day (ld) after reperfusion are shown. Note the increase in HSCT0 mRNA levelsin the ipsilateral(left side in the figure)granulecellsof dentate gyrus (small arrowhead) and hippocampaiCA3 cells (large arrowhead)after transient MCA occlusion.Bar = 1.1 ram.
ltistopathological study Cresyl violet staining of sections of brains at 7 days showed that great cell damage occurred in the caudate (Fig. 5D), whereas cortical cells were barely damaged even at this time (Fig. 5C), as reported previously 2o DISCUSSION Northern blot analysis demonstrated that the HSP70 and HSC70 probes employed in the present study selectively detected HSP70 and HSC70 mRNA, respectively, in rat brain, and did not cross react (Fig. 1). Furthermore the temporal and quantitative changes in HSP70 and HSC70 mRNA levels in Northern blot analysis (Fig. 1) were consistent with that in the in situ hybridization analysis (Figs. 2 and 3). In the in situ hybridization analysis grains for HSP70 and HSC70 probes were located predominantly in the cell body. The specificities of these probes were further supported by the negative signal after hybridization with the irrelevant plasmid vector probe, and by the eliminated signal after the pretreatment with RNAses prior to incubation with the probe for HSPT0 or HSC70 mRNA. These results prove that the HSPT0 and HSC70 eDNA probes used here selectively recognize HSP70 and HSC70 mRNA of rat brain in situ, respectively. HSP70 mRNA was hardly present at caudate and dorsal hippocampal levels of sham animal brain (Fig. 2). On the other hand HSCT0 mRNA was present in most cell populations of rat brain under the conditions of the sham control (Fig. 3). These results suggest different roles for HSP70 and HSC70 under normal conditions, HSC70 has been suggested to associate with nascent polypeptide chains ~,~.4. It is interesting that the distribution of HSC70 mRNA in the sham brain (Fig. 3) was similar to that of total protein synthesis in sham rat 1.9 or gerbil 27,29 brain. Therefore the relatively high level of expression of HSC70 mRNA in cell populations, including CA3 cells, may be related to active protein synthesis in these regions. Clear co.induction of HSP70 and HSCT0 mRNA was observed in cell populations in the MCA territory in post-ischemic brain (Figs. 2 and 3). It is known that HSP70 and HSCT0 proteins bind denatured or unfolded proteins, and degrade or refoid the abnormal proteins, during stressful conditions 15.24. HSP70 is essential for the restoration of normal ribosome assembly, and promotes the synthesis of new ribosomes and proteins and accelerates the recovery of cells 2.~ by ATP.dependent mechanisms ~7. It has been suggested that HSC70 plays a role in the disassembly of the clathrin cage, and participates in post-translational
2OO
Fig. 5. Representativephotographsof cells of cerebralcortex(A and C) and caudate(B and D) in sham(A and B) and post-ischemicbrain at ? days(C and D), Notethe greatamountof cell damagein the caudateat ? daysafter transient local ischemia(D) in contrast to little cell damage in the cortex(C), Bar - 130~,m.
transmembrane targeting of proteins to cellular organelles ~1.Thus the co-induction of HSP70 and H$C70 mRNA indicates a cooperative role for these proteins under such stressful conditions as transient ischemia. This may also suggest an active folding of damaged and/or newly synthesized proteins in the cells of postischemic brain. Despite the general co-induction of HSP70 and
HSC70 mRNA in cells of the MCA territory after ischemia, only HSC70 mRNA was continously induced from the early period of reperfusion in granule cells of the ipsilateral dentate gyrus and hippocampal CA3 cells. However, the CAI cells were never affected (Fig. 4). Recently, in amphetamine.treated rats, Miller et al. a9 observed that HSCT0 mRNA relative levels increased at body temperatures greater than 39°C,
whereas HSPT0 mRNA synthesis was induced at temperatures greater than 40°C, They also suggested that the elevation in HSCT0 mRNA levels would be sufficient to protect cells from lower levels of stress, and that HSP70 would be induced only during higher levels of stress, Therefore our results showing selective induction of HSC70 mRNA in hippocampal cells may indicate that a stress response, which is not accompanied by HSP70 mRNA induction, occurs in the brain after transient ischemia. HSPT0 has been demonstrated to be induced by transient ischemia in rodent brain 14.16,2sand to play a protective role during various stressful conditions of cells 3,(,,z6ag,z~,zs After 30 rain of transient focal ischemia in the animal model used in the present study, as shown in Fig, 5, a great amount of cell damage
201
occurred in the caudate, while cortical cells were almost fully preserved at 7 days 20. Abe et al. z showed that the different susceptibilities of these cells may not be due to the difference in blood flow during ischemia and reperfusion. Therefore it is of interest to notice the temporal and quantitative differences in the maximum induction of HSPT0 mRNA between the cortex and the caudate. The maximum induction of HSP70 mRNA in the caudate was smaller in amount and earlier in peak time than that in the cortex (Fig. 2). These results suggest that the small induction of HSPT0 mRNA at the earlier stage after ischemia may relate to the greater amount of damage to cells in the caudate. In fact HSPT0 mRNA was more greatly and continuously induced in the cortical ceils, which were almost spared even at 7 days. A marked reduction in HSC70 mRNA levels was observed in the caudate at 8 h (Fig. 3) when the small amount of HSP70 mRNA was still elevated (Fig. 2), suggesting possible different transcriptional mechanisms in induction between HSP70 and HSCT0 mRNA in the caudate cells after transient ischemia. It is suspected that HSP70 aids HSC70 in salvaging denatured proteins by solubilizing them and facilitating refolding, or chaperoning them to a degradative system zs. in addition to the cooperative roles with HSP70 as a molecular chaperone in stressful conditions z'~ HSC70 binds the exposed loop of clathrin light chain ~2, and is involved in disassembling the clathrin cage ~. A recent report .~0 indicated that the induction of clathrin might be involved in neuronal death after ischemia. Although a possible change in clathrin was not examined in this experiment the marked reduction of HSC70 mRNA levels in cells of the caudate at the early phase after ischemia may relate to the greater amount of
neuronal cell damage in the caudate, Imbalance of excitatory and inhibitory neuronal innervation and/or differences of cell metabolism have been pointed out which, at least in part, account for the regional differences in susceptibility between the caudate and cortex to ischemia ~. Although further study, such as immunohistochemical analysis, should be performed the spatial changes in the distribution of HSPT0 and HSC70 mRNA detected by in situ hybridization using selective probes suggest roles for HSP70 and HSC70 in the different susceptibilities among cell populations to the transient ischemia.
Acknowledgments. This work was partly supported by Monbusho Grant 01044018 and 03404028, and a grant from the Ministry of Welfare and Health of Japan, and the Kanae foundation. The authors would like to express their appreciation to Dr. M. Aoki and Mrs. M. Matsumoto for their excellent technical assistance.
REFERENCES 1 Abe, K., Araki, T. and Kogure, K., Recovery from edema and of protein synthesis differs between the cortex and caudate following transient focal cerebral ischemia in rats, J. Neurochem., 51 (1988) 1470-1476. 2 Abe, K., Kawagoe, J., Sate, S., Sahara, M. and Kogure, K., Induction of the 'zinc finger' gene after transient focal ischemia in rat cerebral cortex, Neurosci. Lett., 123 (1991) 248-250. 3 Abe, K., Tanzi, R.E. and Kogure, K., Induction of HSPT0 mRNA after transient ischemia in gerbil brain, Neurosci. Lett., 125 (1991) 166-168. 4 Araki, T., Kate, H. and Kogure, K., Selective neuronal vulnerability following transient cerebral ischemia in the gerbil: distribution and time course, Acta Neurol. Scand., 80 (1989) 548-553. 5 Araki, T., Kate, H., Inoue, T. and Kogure, K., Long-term observations on calcium accumulation in post-ischemic gerbil brain, Acta NeuroL Scand., 83 (1991) 244-249. 6 Barbe, M.F., Tytell, M., Gower, D.J. and Welch, W.J., Hyperthermia protects against light damage in the rat retina, Science, 241 (1988) 1817-1820. 7 Beckmann, R.P., Mizzen, L.A. and Welch, W.J., Interaction of Hsp 70 with newly synthesized proteins: implications for protein folding and assembly, Science, 248 (1990) 850-854. 8 Brown, i.R., Rush, S. and Ivy, G.O., Induction of a heat-shock gene at the site of tissue injury in the rat brain, Neuron, 2 (1989) 1559-1564. 9 Dienel, G.A., Pulsinelli, W.A. and Duff, T.E., Regional protein synthesis in rat brain following acute hemispheric ischemia, J. Neurochem., 35 (1980) 1216-1226. 10 Feinberg, A.P. and Vogelstein, B., A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity, AnaL Biochem., 132 (1983) 6-13. I1 FlaherW, K.M., DeLuca-Flaherty, C. and McKay, D.B., Threedimensional structure of the ATPase fragment of a 70 kDa heat-shock cognate protein, Nature, 346 (1990) 623-628. 12 Gething, M.J. and Samhrook, J., Protein folding in the cell, Nature, 355 (1992)33-45. 13 Gonzalez, M.F., Shiraishi, K., Hisanaga, K., Sagar, S.M., Mandabach, M. and Sharp, F,R,, Heat.shock proteins as markers of neuronal injury, MoL Brain Res., 6 (1989)93-100. 14 Gonzalez, M.F., Lowenstein, D., Fernyak, S., Hisanaga, K., Si. mon, R. and Sharp, F.R., Induction of heat.shock protein 72-like lmmunoreactivity in the hippocampal formation following translant global ischemla, Brain Res. Bull,, 26 (1991)241-250. 15 Hightower, L,E,, Heat shock, stress proteins, chaperones, and proteotoxicity, Cell, 66 (1991) 191-19% 16 Kirino, T., Tsujita, Y. and Tamura, A., Induced tolerance to ischemia in gerbil hippocampal neurons, J. eereb. Blood Flow Metab., 11 (1991) 299-307. 17 Lewis, M.T. and Pelham, H.R.B., Involvement of ATP in the nuclear and nucleolar functions of the 70 kDa heat-shock protein, EMBO J., 4 (1985) 3137-3143. 18 Lindquist, S., The heat-shock response, Annu. Rel,. Biochem., 55 (1986) 1151-1191. 19 Miller, E.K., Raese, J.D. and Morrisoh-Bogorad, M., Expression of heat-shock protein 70 and heat-shock cognate 70 messenger RNAs in rat cortex and cerebellum after heat shock or amphetamine treatment, J. Neurochem., 56 (1991) 2060-2071. 20 Nakano, S., Kogure, K. and Fujikura, H., ischemia-induced slowly progressive neuronal damage in the rat brain, Neuroscience, 38 (1990) 115-124. 21 Nowak Jr., T.S., Bond, U. and Schlesinger, iv.~.J., [i~a~ ~hock RNA levels in brain and other tissues after hyperthermla and transient ischemia, J. Neurochem., 54 (1990) 451-458. 22 Nowak Jr., T.S., Localization of 70 kDa stress protein mRNA induction in gerbil brain after ischemia, J. Cereb. Blood Flow Metab., 11 (1991) 432-439. 23 Pelham, H.R.B., HSP70 accelerates the recovery of nucleolar morphology after heat shock, EMBO J., 3 (1984) 3095-3100. 24 Pelham, H.R.B., Functions of the hsp70 protein family: an overview, in R.I. Morimoto, A. Tissi[res and C. Georgopoulos
202 (Eds.)0 Stress Proteins in Biology and Medicine, Cold Spring Harbor Laboratory Press, New York, 1990, pp. 287-299. Riabowol, K.T., Mizzen, L.A. and Welch, W.J., Heat shock is lethal to fibroblasts microinjected with antibodies against hspT0, Science, 242 (1988) 433-436. 26 Sato, S., Abe, K., Kawagoe, J., Saito, A. and Kogure, K., Molecular cloning of heat-shock protein (HSP) 70 gene in post-ischemic gerbil brain, J. Cereb. Blood Flow Metab., 11 (suppl. 2) (1991) 5345. 27 Thilmann, R., Xie, Y., Kleihues, P. and Kiessling, M., Persistent inhibition of protein synthesis precedes delayed neuronal death in ix)st-ischemic gerbil hippocampus, Acta Neuropathol. (Berlin), 71 (1986) 88-93. 28 Vass, K., Welch, W.J. and Nowak Jr., T.S., Localization of
70-kDa stress protein induction in gerbil brain after ischemia, Acta Neuropathol. (Berlin), 77 (1988) 128-135. 29 Yoshidomi, M., Hayashi, T., Abe, K. and Kogure, K., Effects of a new calcium channel blocker, KB-2796, on protein synthesis of the CAI pyramidal cell and delayed neuronal death following transient forebrain ischemia, J. Neurochem., 53 (1989) 1589-1594. 30 Yoshimi, K., Takeda, M., Nishimura, T., Kudo, T., Nakamura, Y., Tada, K. and Iwata, N., An immunohistochemical study of MAP2 and clathrin in gerbil hippocampus after cerebral ischemia, Brain Res., 560 (1991) 149-158. 31 Yoshioka, M., Nagano, l., Nakamura, S., Imaizumi, M. and Kimura, N., Detection of vasoactive intestinal polypeptide messenger RNA in ganglioneuroblastoma by in situ hybridization, Endocrinol. Pathol., 1 (1990)51-57.
195
© 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 17953
Distributions of heat shock protein (HSP) 70 and heat shock cognate protein (HSC) 70 mRNAs after transient focal ischemia in rat brain J. Kawagoe, K. Abe, S. Sato, I. Nagano, S. N a k a m u r a and K. K o g u r e Department of Neurololo, Institute of Brain Diseases, Tohoku University School of Medicine, Sendal (Japan) (Accepted 10 March 1992)
Key words: Heat shock protein; lteat shock cognate protein; Cerebral focal ischemia; in situ hybridization; Rat
The distribution of heat shock protein (HSP) 70 and heat shock cognate protein (HSC) 70 mRNA after 30 min of middle cerebral artery (MCA) occlusion was investigated in rat brain by in situ hybridization using cloned eDNA probes s~.lective for the mRNAs. While HSP70 mRNA was hardly present at caudate and dorsal hippocampal levels of the sham brain this mRNA was greatly induced in cells of the MCA territory ! h after reperfusion. Although the maximum amount of induced HSP70 mRNA in the caudate was much smaller than that in the cortex the maximum induction in the caudate (3 h) preceded that in the cortex (8 h). in contrast to the case of HSP70 mRNA, HSC70 mRNA was present in most cells of the sham brain, and was especially dense in hippocampal CA3 cells. Further induction of HSC70 mRNA was observed after reperfusion in the same cell populations, as in the case of HSP70 mRNA. HSC70 mRNA levels were significantly reduced in the caudate at 8 h when small amounts of HSPT0 mRNA were still elevated. In the ipsilateral granule cells of the dentate gyrus and hippocampal CA3 cells a slight but significant induction of HSC70 mRNA was observed from I h to I day, while obvious induction of HSPT0 mRNA never occurred. All the induced signals of HSPT0 and HSC70 mRNA were diminished or returned to the sham level by 7 days, except for HSC70 mRNA in the caudate. These results are the first observations of the distribution of HSP70 and HSC70 mRNA after transient focal ischemia of rat brain. The results indicate that the temporal and quantitative differences in the maximum induction of HSP70 and HSC70 mRNA may relate to the different susceptibilities between cells of the caudate and cortex. Different roles under normal conditions and a general cooperative role in the recovery process from ischemic injury between HSP70 and itSC7(l are suggested. The finding of selective induction of HSC70 mRNA in hippocampal cells may indicate that a stress response, which is not accompanied by tlSP70 mRNA induction, occurs in the brain after transient ischemia.
INTRODUCTION Stressful conditions of brain, such as tissue injury s, heat shock t~, administration of drugs t.~,t,J and ischemia 2,.~,t.~,t4,t~,a,,aa,2s induce 'stress responses', including the synthesis of the 70 kDa heat-shock protein (HSP70). HSP70 has been suggested to play a protective role under stressful conditions of cells 3.~,.t~,,tg.23.as. Recent studies suggested regional differences in immunoreactivity for HSP70 protein among the hippocampal cell populations corresponded to different vulnerabilities after transient global ischemia in gerbil t¢,,28. These results indicate an important role for HSP70 in protecting cells against injury and/or promoting the recovery of injured cells. The degree of cell damage also differs among cell populations of the ischemic territory after a middle cerebral artery (MCA) occlusion in rat brain 1,20.
Although the distribution of induced HSP70 mRNA has been reported in the global ischemia model of gerbil 22 there has been no similar report in a focal cerebral ischemia model in rat, Thus it is of interest to investigate the change in the spatial distribution of HSP70 mRNA after focal cerebral ischemia in rat for a proposal of a mechanism for the regional differences in cell damage. Heat shock cognate protein (HSC) 70 is a member of the HSP70 family, and is constitutively expressed during normal cell development and differentiation I I The induction of HSC70 mRNA has been investigated under normal and stressful conditions, including ischemia in gerbil and rat brain, by Northern blot analysis 3.tt~.2t. Abe et al. a observed a regional difference in the induction of HSC70 mRNA after transient ischemia, and proposed a possible protective role for
Correspondence: J.-i. Kawagoe, Department of Neurology, Institute of Brain Diseases, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendal 980, Japan. Fax: (81) (22) 272-5818.
196
HSC70 protein in cell damage after ischemia. However, the distribution of HSC70 mRNA in subregions of normal and post-ischemic rat brain has not yet been investigated. We recently isolated eDNA clones selective for HSP70 and HSC70 mRNAs from gerbil brain -'~'.Using these eDNA clones as probes we investigated the spatial distribution of HSP70 and HSC70 mRNA by in situ hybridization in rat brain after 30 min of focal ischemia. MATERIALS AND METHODS
at a concentration of 0.1 n g / # l in prehybridization solution supplemented with 10% dextran sulfate and I0 mM dithiothreitol. The eDNA inserts were radiolabeled with [a -3sS]deoxycytidine triphosphate (I,000 Ci/mmol, Amersham) by random-primer labeling io using a kit (Boehringer Mannheim Yamanouchi Co., Tokyo, Japan), resulting in specific activities of 5 - 6 x 10~ dpm//zg. The control probe was a similarly labeled plasmid vector (pHSG396, Takara Shuzo Co., Kyoto, Japan) fragment (1.1 kb) digested by Hinfl. The sections were washed at room temperature for 2 h in 2 x SSC and for ! h in ! x SSC, and were then dehydrated through a graded ethanol series containing 0.3 M ammonium acetate. After visualization of the hybridization by exposure to X-ray film for 24 h at room temperature the sections were dipped in a liquid emulsion (NR-M2, Konica Co., Tokyo, Japan) diluted (1:1) with 0.6 M ammonium acetate, They were then exposed at 4°C for approximately 3 weeks, developed, and counterstained with hematoxylin. Several slides were treated with 100 p,g/ml RNAse A and 10 U / m l RNAse TI (Sigma, St. Louis, MO, USA) in 2 x SSC at 37"C for 2 h prior to prehybridization, and were then hybridized with the probe for HSPT0 or HSCT0 mRNA,
.4nunal mo&'l The animal model used in the present study consisted of a middle cerebral artery (MCA) occlusion using a microembolus of nylon thread i,.'. Male Wistar rats weighing 2511-2811 g were lightly anesthetized by inhalation of a nitrous oxide/oxygen/halothane (6t)c; : 31|¢; : If; ) mixture during `surgical preparation. The right MCA wa`s expo`sed and then the anesthesia was stopped. When the animal began to regain consciousness the origin of the artery was occluded h)r 311 min by insertion of a nylon thread via the common carotid artery. Body temperatare was monitored in :dl animals with a rectal probe, and was maintained at 37°C using a heat pad daring the `surgical preparation and occlusion of the MCA. After 30 rain of hchemia the bh)od flow was restored by removal of the nylon thread, and no further attempt was made to maintain constant body temperalure of ,mimah, The animals were allowed It) recover for I, 3. 8 h, and I, 2 and 7 days (n - 3 for each time point) at ambient temper:ilure (21=23°C), and then were de~:apitated, Sham control animals wer~ treated with cervical ,~urtlery but without the followin~ insertion of nyh)n thread.
Histopathologicai study Ischemia-induced cell damage was estimated at 7 days by light microscopic observation of Cresyl violet-stained sections (5/~m) of brains receiving conventional fixation 4,s.
RESULTS The results obtained in hybridization and histopathological studies were reproducible in all animals employed at each time point.
Hybridization experiments Figure ! shows the results of Northern blot analysis using cloned HSP70 and HSC70 eDNA probes against
i b.t)ndi:atiot) (',~7)('rim('nr~ The insem of eh),ed ~DNAs used a,~ probes in the present study weru originally from th~ eerd)r.l cortex t)l' ~erbil, The ~izes of the inserts `selective I'or tlSP711 and HSC711 mRNA were 1.11(pGA ~)and 1.4 kb (pClD~), respectively ~r, The ~pe¢ifi¢ities of these probes for lISP70 and t!SC7(1 mRNA in rat brain were examined by Northern blot analy,~is. Total RNA was extracted from cerebral cortices of the t)c¢luded MCA territory, electrophoresed, and transferred to a nylon membrane, as in our previous report ~, The probes were radiolabeled with [ , - ~ " P ] dATP 16,01H1 Ci/mmol, Amersham, Tokyo, Japan) by random-primer labeling, and hybridized against Northern Idols 115/~g of total RNA/lane) at 42°C for 20 h in a hybridization solution containing formamide, After the hybridization the filters were wa`shed with 2×SSC (I ×SSC - 151)mM NaCI, 15 mM sodium citrate), I x SSC, and 0,5 x SSC with 11,2~ SDS at n)om temperature t)r t)5°C. The filters were exposed to X-ray film for 21) h at -8(I°C, In situ hybridization was performed by the method of Yoshioka el al. -~) with a slight modification. Briefly, the dissected brains were frozen in powdered dry ice and stored at - 800C, Sections ( Ill/,tin) at the caudate and dorsal hippocampal levels were cut on a cryostat at -18°C, and collected onto slides coated with Histostik (Accurate Chemical and Scientific Corp,, We`stbury, NY), The sections were fixed h)r 11) rain in an ethanol/acetic acid (3:1) mixture, transferred to 11.2 M HCI for 211min, and immersed tilt 211rain at 5II°C in 2 x SSC (pH 7.1)). They were then digested h)r 15 rain at 37°C with Hill ptg/ml of proteinase K (Merk & Co,, Rahway, NJ) in 21) mM Tris-HCI buffer (pH 7.4)with 2 raM CaCI,, and dehydrated through a graded ethanul ,series, Slides were prehybridized for 2 h at room temperature in a solution containing 511% tbrmamide, 81~11mM NaCI, IU mM Tris-HCI (pH 7.5), I mM EDTA, 0,02% polyvinylpyrrolidone, (),U2r~ Ficoll, I),02Ci bovine serum albumin, I(X) p,g/ml sonicated, denatured calf thymus DNA, and 1),5 mg/ml calf liver RNA, Hybridization was done for 21) h at 42°C using ~~S-labeled eDNA probe
HSP 70
HSC 70
S
S 8h 2d
8h 2d
Fig, I, Northern blot analysis of HSPT0 and HSC70 mRNAs in rat cerebral cortices of the ipsilateral MCA territory of sham control (S) and post-ischemic brain at 8 h (8h) and 2 days (2d) after the reperfusion. Note that HSPT0 and HSCT0 probes detected the different mRNA species in the rat brain and did not cross react, Small and large arrowheads show positions of the detected HSPT0 (2,8 and 3,0 kb) and HSC70 (2.4 kb) mRNAs, respectively.
197 total RNA isolated from rat cerebral cortex in the present ischemic model. Figure 1 shows the autoradiogram of the membrane filter finally washed with 0.5 x SSC at 65°C. A final wash with 2 or 1 x SSC at room temperature showed the same results (data not shown). In the sham control brain there was no hybridization of HSP70 mRNA, whereas HSC70 mRNA was identified. At 8 h after repeffusion HSP70 mRNA levels were greatly induced, and a slight increase in the HSC70 mRNA level was also observed. Thus HSP70 and HSC70 probes used in the present study selectively
detected the HSP70 mRNA doublet (2.8 and 3.0 kb) and the HSC70 mRNA species (2.4 kb), respectively, in rat brain. The identification patterns of HSP70 and HSC70 mRNAs with eDNA probes (Fig. l) were similar to that in the previous report of Northern blot analysis using oligonucleotide probes specific for HSPT0 and HSC70 mRNAs ~9. By 2 days the amount of HSPT0 mRNA had decreased, and that of HSC70 mRNA had returned to almost the sham control level. Neither the HSP70 probe nor the HSC70 probe cross-hybridized with any other mRNA species.
S
/ •
I h ~
3h
•
~
~,
, •
•
.
'.,,~_" "
~'.t' .,"
..
8h
(,~, '
~.
ld ~,~
,,
.(.,
2d
d
' f"
b ¸
m Fig. 2. Distribution of HSPT0 mRNA at caudate (left column) and dorsal hippocampal (right column) levels of sham (S) and post.ischemic rat brain. Bar = 2 ram.
198 Distribution of HSP70 mRNA in post-ischemic rat brain at caudate and dorsal hippocampal levels is shown in Fig. 2. While HSP70 mRNA was hardly present in the sham brain the mRNA was induced in cells of the caudate and cerebral cortex even 1 h after reperfusion. The HSP70 mRNA induction was gradually intensified, with peaks at 3 h for cells of the caudate and 8 h for the cortex. The maximum amount of induced HSP70 mRNA in the caudate was smaller than that in the cortex, in the caudate the induced HSPT0 mRNA level
was greatly reduced at 8 h, and the hybridization signal was almost diminished at 1 day. All of the induced signal of HSP70 mRNA disappeared by 7 days after reperfusion. Spatial changes of distribution of HSC70 mRNA in sham and post-ischemic rat brain at caudate and dorsal hippocampai levels are documented in Fig. 3. In contrast to HSP70 mRNA, HSC70 mRNA was present in most cell populations of the sham brain. A relatively high level of HSC70 mRNA was observed in cells of
S
lh
3h
8h d I
ld
2d
7d
m
Fig, 3. Distribution of H$C70 mRNA at caudate (left column) and dorsal hippocampal (right column) levels of sham ($) and post-ischemic rat brain. Bar = 2 ram,
199 the medial habenula and hippocampal CA3. The transient focal ischemia increased the amount of HSC70 mRNA in the same cell populations as in HSPT0 mRNA, with the same peak times. In addition to those cell populations HSC70 mRNA was induced in the ipsilateral granule cells of dentate gyrus and hippocampal CA3 cells from 1 h to I day (Figs. 3 and 4). The magnitude of HSC70 mRNA induction was greater in dentate granule cells than CA3 cells, however, CA1 cells were never affected. At 8 h after reperfusion HSC70 mRNA levels were greatly reduced in the caudate (Fig. 3) while a small amount of HSP70 mRNA was still elevated (Fig. 2). The significant reduction of HSC70 mRNA levels in the caudate continued to 7 days. The induced HSC70 mRNA returned to the sham level by 7 days except for the caudate. Observation of the sections dipped in the liquid emulsion revealed that grains for HSP70 and HSC70 probes were predominantly located in the cell body (data not shown). Neither the sections incubated with plasmid vector probe nor the sections treated with RNAses before incubation with the probe for HSP70 or HSC70 mRNA exhibited any specific hybridization (data not shown).
HSC70
S lh ld Fig. 4. High magnificationphotographsof the spatial change in the distribution of HSCT0mRNA at the dorsal hippocampallevel. The sections of sham (S) and post-ischemicbrain at I h (lh) and 1 day (ld) after reperfusion are shown. Note the increase in HSCT0 mRNA levelsin the ipsilateral(left side in the figure)granulecellsof dentate gyrus (small arrowhead) and hippocampaiCA3 cells (large arrowhead)after transient MCA occlusion.Bar = 1.1 ram.
ltistopathological study Cresyl violet staining of sections of brains at 7 days showed that great cell damage occurred in the caudate (Fig. 5D), whereas cortical cells were barely damaged even at this time (Fig. 5C), as reported previously 2o DISCUSSION Northern blot analysis demonstrated that the HSP70 and HSC70 probes employed in the present study selectively detected HSP70 and HSC70 mRNA, respectively, in rat brain, and did not cross react (Fig. 1). Furthermore the temporal and quantitative changes in HSP70 and HSC70 mRNA levels in Northern blot analysis (Fig. 1) were consistent with that in the in situ hybridization analysis (Figs. 2 and 3). In the in situ hybridization analysis grains for HSP70 and HSC70 probes were located predominantly in the cell body. The specificities of these probes were further supported by the negative signal after hybridization with the irrelevant plasmid vector probe, and by the eliminated signal after the pretreatment with RNAses prior to incubation with the probe for HSPT0 or HSC70 mRNA. These results prove that the HSPT0 and HSC70 eDNA probes used here selectively recognize HSP70 and HSC70 mRNA of rat brain in situ, respectively. HSP70 mRNA was hardly present at caudate and dorsal hippocampal levels of sham animal brain (Fig. 2). On the other hand HSCT0 mRNA was present in most cell populations of rat brain under the conditions of the sham control (Fig. 3). These results suggest different roles for HSP70 and HSC70 under normal conditions, HSC70 has been suggested to associate with nascent polypeptide chains ~,~.4. It is interesting that the distribution of HSC70 mRNA in the sham brain (Fig. 3) was similar to that of total protein synthesis in sham rat 1.9 or gerbil 27,29 brain. Therefore the relatively high level of expression of HSC70 mRNA in cell populations, including CA3 cells, may be related to active protein synthesis in these regions. Clear co.induction of HSP70 and HSCT0 mRNA was observed in cell populations in the MCA territory in post-ischemic brain (Figs. 2 and 3). It is known that HSP70 and HSCT0 proteins bind denatured or unfolded proteins, and degrade or refoid the abnormal proteins, during stressful conditions 15.24. HSP70 is essential for the restoration of normal ribosome assembly, and promotes the synthesis of new ribosomes and proteins and accelerates the recovery of cells 2.~ by ATP.dependent mechanisms ~7. It has been suggested that HSC70 plays a role in the disassembly of the clathrin cage, and participates in post-translational
2OO
Fig. 5. Representativephotographsof cells of cerebralcortex(A and C) and caudate(B and D) in sham(A and B) and post-ischemicbrain at ? days(C and D), Notethe greatamountof cell damagein the caudateat ? daysafter transient local ischemia(D) in contrast to little cell damage in the cortex(C), Bar - 130~,m.
transmembrane targeting of proteins to cellular organelles ~1.Thus the co-induction of HSP70 and H$C70 mRNA indicates a cooperative role for these proteins under such stressful conditions as transient ischemia. This may also suggest an active folding of damaged and/or newly synthesized proteins in the cells of postischemic brain. Despite the general co-induction of HSP70 and
HSC70 mRNA in cells of the MCA territory after ischemia, only HSC70 mRNA was continously induced from the early period of reperfusion in granule cells of the ipsilateral dentate gyrus and hippocampal CA3 cells. However, the CAI cells were never affected (Fig. 4). Recently, in amphetamine.treated rats, Miller et al. a9 observed that HSCT0 mRNA relative levels increased at body temperatures greater than 39°C,
whereas HSPT0 mRNA synthesis was induced at temperatures greater than 40°C, They also suggested that the elevation in HSCT0 mRNA levels would be sufficient to protect cells from lower levels of stress, and that HSP70 would be induced only during higher levels of stress, Therefore our results showing selective induction of HSC70 mRNA in hippocampal cells may indicate that a stress response, which is not accompanied by HSP70 mRNA induction, occurs in the brain after transient ischemia. HSPT0 has been demonstrated to be induced by transient ischemia in rodent brain 14.16,2sand to play a protective role during various stressful conditions of cells 3,(,,z6ag,z~,zs After 30 rain of transient focal ischemia in the animal model used in the present study, as shown in Fig, 5, a great amount of cell damage
201
occurred in the caudate, while cortical cells were almost fully preserved at 7 days 20. Abe et al. z showed that the different susceptibilities of these cells may not be due to the difference in blood flow during ischemia and reperfusion. Therefore it is of interest to notice the temporal and quantitative differences in the maximum induction of HSPT0 mRNA between the cortex and the caudate. The maximum induction of HSP70 mRNA in the caudate was smaller in amount and earlier in peak time than that in the cortex (Fig. 2). These results suggest that the small induction of HSPT0 mRNA at the earlier stage after ischemia may relate to the greater amount of damage to cells in the caudate. In fact HSPT0 mRNA was more greatly and continuously induced in the cortical ceils, which were almost spared even at 7 days. A marked reduction in HSC70 mRNA levels was observed in the caudate at 8 h (Fig. 3) when the small amount of HSP70 mRNA was still elevated (Fig. 2), suggesting possible different transcriptional mechanisms in induction between HSP70 and HSCT0 mRNA in the caudate cells after transient ischemia. It is suspected that HSP70 aids HSC70 in salvaging denatured proteins by solubilizing them and facilitating refolding, or chaperoning them to a degradative system zs. in addition to the cooperative roles with HSP70 as a molecular chaperone in stressful conditions z'~ HSC70 binds the exposed loop of clathrin light chain ~2, and is involved in disassembling the clathrin cage ~. A recent report .~0 indicated that the induction of clathrin might be involved in neuronal death after ischemia. Although a possible change in clathrin was not examined in this experiment the marked reduction of HSC70 mRNA levels in cells of the caudate at the early phase after ischemia may relate to the greater amount of
neuronal cell damage in the caudate, Imbalance of excitatory and inhibitory neuronal innervation and/or differences of cell metabolism have been pointed out which, at least in part, account for the regional differences in susceptibility between the caudate and cortex to ischemia ~. Although further study, such as immunohistochemical analysis, should be performed the spatial changes in the distribution of HSPT0 and HSC70 mRNA detected by in situ hybridization using selective probes suggest roles for HSP70 and HSC70 in the different susceptibilities among cell populations to the transient ischemia.
Acknowledgments. This work was partly supported by Monbusho Grant 01044018 and 03404028, and a grant from the Ministry of Welfare and Health of Japan, and the Kanae foundation. The authors would like to express their appreciation to Dr. M. Aoki and Mrs. M. Matsumoto for their excellent technical assistance.
REFERENCES 1 Abe, K., Araki, T. and Kogure, K., Recovery from edema and of protein synthesis differs between the cortex and caudate following transient focal cerebral ischemia in rats, J. Neurochem., 51 (1988) 1470-1476. 2 Abe, K., Kawagoe, J., Sate, S., Sahara, M. and Kogure, K., Induction of the 'zinc finger' gene after transient focal ischemia in rat cerebral cortex, Neurosci. Lett., 123 (1991) 248-250. 3 Abe, K., Tanzi, R.E. and Kogure, K., Induction of HSPT0 mRNA after transient ischemia in gerbil brain, Neurosci. Lett., 125 (1991) 166-168. 4 Araki, T., Kate, H. and Kogure, K., Selective neuronal vulnerability following transient cerebral ischemia in the gerbil: distribution and time course, Acta Neurol. Scand., 80 (1989) 548-553. 5 Araki, T., Kate, H., Inoue, T. and Kogure, K., Long-term observations on calcium accumulation in post-ischemic gerbil brain, Acta NeuroL Scand., 83 (1991) 244-249. 6 Barbe, M.F., Tytell, M., Gower, D.J. and Welch, W.J., Hyperthermia protects against light damage in the rat retina, Science, 241 (1988) 1817-1820. 7 Beckmann, R.P., Mizzen, L.A. and Welch, W.J., Interaction of Hsp 70 with newly synthesized proteins: implications for protein folding and assembly, Science, 248 (1990) 850-854. 8 Brown, i.R., Rush, S. and Ivy, G.O., Induction of a heat-shock gene at the site of tissue injury in the rat brain, Neuron, 2 (1989) 1559-1564. 9 Dienel, G.A., Pulsinelli, W.A. and Duff, T.E., Regional protein synthesis in rat brain following acute hemispheric ischemia, J. Neurochem., 35 (1980) 1216-1226. 10 Feinberg, A.P. and Vogelstein, B., A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity, AnaL Biochem., 132 (1983) 6-13. I1 FlaherW, K.M., DeLuca-Flaherty, C. and McKay, D.B., Threedimensional structure of the ATPase fragment of a 70 kDa heat-shock cognate protein, Nature, 346 (1990) 623-628. 12 Gething, M.J. and Samhrook, J., Protein folding in the cell, Nature, 355 (1992)33-45. 13 Gonzalez, M.F., Shiraishi, K., Hisanaga, K., Sagar, S.M., Mandabach, M. and Sharp, F,R,, Heat.shock proteins as markers of neuronal injury, MoL Brain Res., 6 (1989)93-100. 14 Gonzalez, M.F., Lowenstein, D., Fernyak, S., Hisanaga, K., Si. mon, R. and Sharp, F.R., Induction of heat.shock protein 72-like lmmunoreactivity in the hippocampal formation following translant global ischemla, Brain Res. Bull,, 26 (1991)241-250. 15 Hightower, L,E,, Heat shock, stress proteins, chaperones, and proteotoxicity, Cell, 66 (1991) 191-19% 16 Kirino, T., Tsujita, Y. and Tamura, A., Induced tolerance to ischemia in gerbil hippocampal neurons, J. eereb. Blood Flow Metab., 11 (1991) 299-307. 17 Lewis, M.T. and Pelham, H.R.B., Involvement of ATP in the nuclear and nucleolar functions of the 70 kDa heat-shock protein, EMBO J., 4 (1985) 3137-3143. 18 Lindquist, S., The heat-shock response, Annu. Rel,. Biochem., 55 (1986) 1151-1191. 19 Miller, E.K., Raese, J.D. and Morrisoh-Bogorad, M., Expression of heat-shock protein 70 and heat-shock cognate 70 messenger RNAs in rat cortex and cerebellum after heat shock or amphetamine treatment, J. Neurochem., 56 (1991) 2060-2071. 20 Nakano, S., Kogure, K. and Fujikura, H., ischemia-induced slowly progressive neuronal damage in the rat brain, Neuroscience, 38 (1990) 115-124. 21 Nowak Jr., T.S., Bond, U. and Schlesinger, iv.~.J., [i~a~ ~hock RNA levels in brain and other tissues after hyperthermla and transient ischemia, J. Neurochem., 54 (1990) 451-458. 22 Nowak Jr., T.S., Localization of 70 kDa stress protein mRNA induction in gerbil brain after ischemia, J. Cereb. Blood Flow Metab., 11 (1991) 432-439. 23 Pelham, H.R.B., HSP70 accelerates the recovery of nucleolar morphology after heat shock, EMBO J., 3 (1984) 3095-3100. 24 Pelham, H.R.B., Functions of the hsp70 protein family: an overview, in R.I. Morimoto, A. Tissi[res and C. Georgopoulos
202 (Eds.)0 Stress Proteins in Biology and Medicine, Cold Spring Harbor Laboratory Press, New York, 1990, pp. 287-299. Riabowol, K.T., Mizzen, L.A. and Welch, W.J., Heat shock is lethal to fibroblasts microinjected with antibodies against hspT0, Science, 242 (1988) 433-436. 26 Sato, S., Abe, K., Kawagoe, J., Saito, A. and Kogure, K., Molecular cloning of heat-shock protein (HSP) 70 gene in post-ischemic gerbil brain, J. Cereb. Blood Flow Metab., 11 (suppl. 2) (1991) 5345. 27 Thilmann, R., Xie, Y., Kleihues, P. and Kiessling, M., Persistent inhibition of protein synthesis precedes delayed neuronal death in ix)st-ischemic gerbil hippocampus, Acta Neuropathol. (Berlin), 71 (1986) 88-93. 28 Vass, K., Welch, W.J. and Nowak Jr., T.S., Localization of
70-kDa stress protein induction in gerbil brain after ischemia, Acta Neuropathol. (Berlin), 77 (1988) 128-135. 29 Yoshidomi, M., Hayashi, T., Abe, K. and Kogure, K., Effects of a new calcium channel blocker, KB-2796, on protein synthesis of the CAI pyramidal cell and delayed neuronal death following transient forebrain ischemia, J. Neurochem., 53 (1989) 1589-1594. 30 Yoshimi, K., Takeda, M., Nishimura, T., Kudo, T., Nakamura, Y., Tada, K. and Iwata, N., An immunohistochemical study of MAP2 and clathrin in gerbil hippocampus after cerebral ischemia, Brain Res., 560 (1991) 149-158. 31 Yoshioka, M., Nagano, l., Nakamura, S., Imaizumi, M. and Kimura, N., Detection of vasoactive intestinal polypeptide messenger RNA in ganglioneuroblastoma by in situ hybridization, Endocrinol. Pathol., 1 (1990)51-57.
