Expression of Heat Shock Protein 70 and Heat Shock Cognate 70 ...

Journal of Neurochemistry Raven Press, Ltd., New York 0 1991 International Society for Neurochemistry

Expression of Heat Shock Protein 70 and Heat Shock Cognate 70 Messenger RNAs in Rat Cortex and Cerebellum After Heat Shock or Amphetamine Treatment *t$E. Katherine Miller, $§Joachim D. Raese, and "TMarcelle Morrison-Bogorad Departments of "Neurology, ?Biochemistry, and $Psychiatry, University of Texas South western Medical Center, and §Schizophrenia Research Center, Veterans Administration Medical Center, Dallas, Texas, U.S.A.

Abstract: The expression of strictly inducible hsp70 mRNAs and constitutively expressed hsc70 mRNAs was compared in cerebellum and cerebral cortex of control rats, heat-shocked rats, and rats made hyperthermic with amphetamine. An hsc70-specific oligonucleotide probe identified a 2.55-kb mRNA in cerebellum and cerebral cortex of all rats. An hsp70-specific oligonucleotide probe identified a 3.05-kb mRNA and a 3.53-kb mRNA in cerebellum and cerebral cortex of heat-shocked and amphetamine-treated rats, but not in control rats. Quantitation demonstrated that both Asp70 and hsc70 mRNA levels, relative to 18s rRNA levels, were increased following each treatment. The relative levels of both mRNAs were higher in cerebellum than in cerebral cortex. In amphetamine-treated rats, hsc70 mRNA relative levels increased at body temperatures greater than 39"C,

whereas hsp70 mRNA synthesis was induced at temperatures greater than 40°C. Total thermal response values and relative levels of both mRNAs were compared. The results suggested that both the transcription and turnover of Asp70 mRNAs differed between cerebellum and cerebral cortex. At equivalent total thermal response values, amphetamine-treated rats had higher relative levels of hsp70 mRNAs than heat-shocked rats, suggesting that amphetamine enhanced the induction of hsp70 rnRNAs. Key Words: hsc70 mRNA-hsp70 mRNA-Heat shock-Amphetamine-mRNA inductionRat brain regions. Miller E. K. et al. 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, 2060-207 1 (1 99 1 ).

In all organisms studied, heat shock and a number of other stresses induce the heat shock response. This response is characterized by the rapid transcription and translation of several molecular weight families of heat shock genes (for reviews, see Schlesinger et al., 1982; Craig, 1985, 1989; Lindquist, 1986; Pelham, 1986, 1989; Lindquist and Craig, 1988; Morimoto et al., 1990).These different heat shock proteins may function to protect cells from the adverse consequences of heat and other stresses. Several studies, in a variety of systems from yeast to rat retina (McAlister and Finkelstein, 1980; Barbe et al., 1988; Johnston and Kucey, 1988; Riabowol et al., 1988; Chopp et al., 1989; Magnusson and Wieloch, 1989), have shown a correlation between the presence of induced heat shock proteins and the ability of cells to survive following stress. Among the most abundant of the heat shock proteins

is a family of highly conserved proteins approximately 70 kDa in size. Members of this family function as chaperones in a wide variety of cellular processes, primarily by maintaining proteins in unfolded conformations necessary for transport, translocation, or assembly (for reviews, see Pelham, 1986, 1989; Rothman and Schmid, 1986; Deshaies et al., 1988; Craig, 1989; Ellis and Hemmingsen, 1989;Sambrook and Gething, 1989; Morimoto et at., 1990; Schlesinger, 1990). The strictly heat-inducible member of this family, heat shock protein 70, is not expressed under normal physiological conditions, but is induced by heat shock and other stresses (Lindquist and Craig, 1988).Several of the strictly inducible heat shock protein 70 genes (hsp70 genes) or their cDNAs have been cloned and characterized in mammals (Lowe and Moran, 1984; Voellmy et al., 1985; Leung et al., 1990). A closely

Received July 24, 1990; revised manuscript received December 3. 1990; accepted December 5 , 1990. Address correspondence and reprint requests to Dr. M. MomsonBogorad at Department of Neurology, University of Texas Southwestern Medical Center, 5323 Hany Hines Blvd., Dallas, TX 75235, U.S.A.

Abbreviations used: LSD, lysergic acid diethylamide; SDS, sodium dodecyl sulfate; SSC, standard sodium citrate; SSPE, standard sodium phosphate EDTA; TTR, total thermal response.

2060

HEAT SHOCK PROTEIN 70 mRNA EXPRESSION IN RAT BRAIN related member of the human hsp70 gene family is the primate-specific heat shock protein 70 X gene (hsx70 gene; Hunt and Morimoto, 1985). The hsx70 mRNA has been identified in a number of cell lines of primate origin (Pelham, 1986). The hsx70 gene is constitutively expressed, cell cycle-regulated, and strongly induced by serum treatment and heat shock (Wu and Morimoto, 1985). The heat shock cognate 70 genes (hsc70 genes) are members of this multigene family that are constitutively expressed at normal temperatures and found in both primates and nonpnmates. Several mammalian hsc70 genes have been cloned and sequenced (O’Malley et al., 1985; Mues et al., 1986; Dworniczak and Mirault, 1987; Giebel et al., 1987; Sorger and Pelham, 1987). Like the primate-specific hsx70 gene, the hsc70 genes are cell cycle-regulated (Morange et al., 1984; Sorger and Pelham, 1987). However, they are induced only 1.5- to twofold by hyperthermia in tissue culture (O’Malley et al., 1985; Sorger and Pelham, 1987; Miller et al., unpublished observations). The different members of the hsp70 gene family have a high level of intrafamily sequence conservation (Hunt and Morimoto, 1985; Lindquist and Craig, 1988). Consequently, nucleic acid probes containing heat shock coding sequences often do not distinguish between inducible and constitutively synthesized mRNAs (Sprang and Brown, 1987; Brown et al., 1989; Nowak et al., 1990). Further, “hsp70” is often used to denote both the strictly inducible and the constitutive members of this gene family. In this article, “hsp70” refers only to the inducible genes or their mRNAs and “hsc70” refers only to the constitutive genes or their mRNAs. In mammalian brain, hsp70 mRNA is induced following a number of physical and chemical stressors. These include heat shock, hypoxia, mechanical trauma, and administration of drugs, such as lysergic acid diethylamide (LSD) and amphetamine, that elicit a hyperthermic response (Brown et al., 1982, 1985, 1989; Brown, 1983; Miller et al., 1987; Uney et al., 1988; Nowak et al., 1990).Early studies in Aplysiu (Greenberg and Lasek, 1985) and in mouse (White, 1980) show that heat shock protein 70 was induced most strongly following heat shock in nonneuronal cells. In situ hybridization studies, with a cDNA probe hybridizing to both hsp70 and hsc70 mRNAs, show that heat shock protein 70-related mRNAs are most abundant in neurons of control rabbit and rat brain (Sprang and Brown, 1987; Brown et al., 1989; Masing and Brown, 1989). Highest levels of heat shock protein 70-related mRNAs are found in glia following LSD administration or heat shock (Sprang and Brown, 1987; Masing and Brown, 1989). In these studies, probes hybridized to both hsp70 mRNAs and hsc70 mRNAs, making it impossible to determine whether the steady-state levels of the constitutively synthesized hsc70 mRNAs are altered during drug-induced hyperthermia or heat shock. In a more recent study, Brown and Rush (1990) used probes derived from the noncoding regions of the inducible and

2061

constitutively expressed members of the hsp70 gene family to show the time course of induction of Asp70 mRNAs following hyperthermia. Their northern analysis showed little or no increase in the levels of the hsc70 mRNAs after hyperthermia (Brown and Rush, 1990). In this study, we examined hsc70 mRNA and hsp70 mRNA levels in the cerebral cortex and cerebellum of rat brain following amphetamine-induced hyperthermia or heat shock treatment. We found that levels of both mRNAs were increased following a sufficient degree of hyperthermia. The temperatures required for hsp70 and hsc70 mRNA induction were different in amphetamine-treated rats. The time courses and the extent of induction of both mRNAs were different in the two brain regions and in the two treatments. EXPERIMENTAL PROCEDURES Amphetamine treatment Adult (250 g) female Sprague-Dawley rats were injected intraperitoneally with either a 10 or 15 mg/kg body weight dose of amphetamine sulfate (Sigma). Control rats were injected intraperitoneally with an equivalent volume of sterile normal saline. The body temperature of each rat was monitored with a thermistor rectal probe (Yellow Springs Instruments 423) connected to a telethermometer (Yellow Springs Instruments 44TD) and recorded at 10-min intervals. All animals were anesthetized with C 0 2gas and killed by decapitation. Brain regions were dissected, frozen in liquid nitrogen, and stored at -80°C.

Heat shock Adult (250 g) female Sprague-Dawley rats, both heatshocked and control, were anesthetized with 0.2 ml of 50 mg/ml sodium Nembutal (Abbott). Rats were heat-shocked with an infrared heat lamp (General Electric) positioned 612 inches above the animal. The body temperature of each rat was monitored continuously. The rate of body temperature increase and maximum temperature attained were controlled by raising or lowering the lamp as necessary. In the first study, rats were rapidly heat-shocked so that their body temperatures increased to approximately 42°C within 15 min. The maximum length of time for heat shock and recovery was 90 min. In the second study, animals were heatshocked more slowly so that their body temperature-versustime profiles paralleled the mean body temperature-versustime profiles of rats administered 15 mg/kg amphetamine. The maximum length of time for heat shock and recovery was 190 min. All animals were killed by decapitation. Brain regions were dissected and processed as described for amphetamine-treated rats. In order to compare rectal temperature with brain temperature, brain temperature immediately post mortem was measured in several heat-shocked and amphetamine-treated rats using a thermistor fast-response surface probe (Yellow Springs Instruments 427). Surface brain temperature never differed by more than 0.5 “C from rectal temperature.

Isolation of total RNA The microisolation method of Ilaria et al. (1985) was used for total RNA isolations with the following modifications. Tissues were placed in 6 M guanidinium HCl (Ultrapure,

J. Neurochem.. Vol. 56, No. 6. 1991

2062

E. K. MILLER ET AL.

BRL), 20 m M sodium acetate, and 20 mM dithiothreitol (pH 5.0) and homogenized on ice for 60 s using a Tissuemizer (Tekmar SDT 1810) with the small probe attachment (Tekmar 8N). RNA concentrations were determined by absorption at 260 nm. A 2-pg sample of each RNA isolate was electrophoresed through a 1% denaturing agarose gel (Maniatis et al., 1982) to assess RNA integrity following isolation.

Oligonucleotide probes The 18s rRNA oligonucleotide probe (Lin and MomsonBogorad, 1990) was complementary to nucleotides 50-73 of the rat 18s rRNA sequence (Torczynski et al., 1983; Berent et al., 1985). The hsp70- and hsc70-specific oligonucleotide probes were selected using the strategy described in Results. Probe sequences were checked against known sequences in both Genbank and EMBL Nucleic Acid Databases. The QUEST and FASTdb programs in the Intelligenetics Suite (Bionet) were utilized to search for sequences either identical to or similar to the selected probes. All oligonucleotides were synthesized on a Biosearch 8600 automated DNA synthesizer. The hsc70, hsp70, and 18s rRNA oligonucleotide probes TP and polynuclewere 5' end labelled with [ Y - ~ ~ P ] A(NEN) otide T4 kinase (Pharmacia) (Maniatis et al., 1982).The hsc70 and hsp70 probes were labelled to a specific activity of 2-4 X lo9 cpm/pg, and the 18s rRNA probe was labelled to a specific activity of 2-4 X lo6 cpm/pg. Radionucleotides unincorporated during the 5' end labelling reactions were removed by cetylpyridinium bromide precipitation (Geck and Nasz, 1983).

Northern blot analysis The standard solutions for hybridization were used and are listed with their abbreviations: (a) 1 X standard sodium citrate (SSC) contained 0.15 M NaCl and 15 mM sodium citrate (pH 7.2); (b) 1 X standard sodium phosphate EDTA (SSPE) contained 180 m M NaCl, 10 mM Na1.5P04,and 1 mM Na2EDTA (pH 8.0); and (c) 1 X Denhardt's solution (Denhardt's) contained 0.02% bovine serum albumin, 0.02% polyvinylpyrrolidone, and 0.02% Ficoll (Denhardt, 1966). Ten micrograms of total RNA were electrophoresed through 1% denaturing agarose gets (Maniatis et al., 1982). RNA was transferred to Zeta-Probe membranes (Bio-Rad) by electroblot transfer according to the Bio-Rad protocol (Bio-Rad, 1982). Vacuumdried membranes were prewashed at 65°C in 0.5% sodium dodecyl sulfate (SDS) and 0.1 X SSC for 4 h. Membranes were prehybridized in 30 ml of 10 X Denhardt's, 1% SDS, 5 X SSPE, 0.1% NaPPi, and 0.1 mg/ ml poly(A) (Sigma) for 8 h at 45°C. Membranes were hybridized to either the [32P]hsp70probe, the [32P]hsc70probe, or the [32P]18SrRNA probe for 12-16 h at 47°C in 30 ml of 5 X Denhardt's, 5 X SSPE, 1% SDS, 0.1% NaPPi, and 0.1 mg of poly(A)/ml. The final inputs of the [32P]hsc70and [32P]hsp70probes were 1-2 X lo7cpm/30 ml of hybridization solution, and the final amount of the [32P]18s rRNA probe was 2-4 X lo3 cpm/30 ml of hybridization solution. The unhybridized excess of [32P]hsc70and [32P]hsp70probes was removed with the following sets of washes: set 1 consisted of three 15-min washes in 6 X SSC, 0.1% SDS, and 0.05% NaPPi at room temperature (22-23°C); set 2 consisted of three 15min washes in 6 X SSC, 0.1% SDS, and 0.05% NaPPi at 47°C; and set 3 consisted of three 15-min washes in 1 X SSC and 0.05% NaPP; at 55-57°C. Membranes hybridized with the ["P]l8S rRNA probe were washed in an identical manner, with the exception that the set 3 washes were done at 50°C. Standard autoradiographic methods were used to expose the J . Neurochem.. Vol. 56, No. 6. 1991

membranes to Kodak XAR 2 x-ray film (Maniatis et al., 1982).

Determination of the relative amounts of hsp70 and hsc70 mRNAs Two- and four-microgram amounts of total RNA for slot blot quantitation were denatured for 15 rnin at 65°C in 7.4% formaldehyde (wt/vol) and 4 X SSC. Each sample was adjusted to 18 X SSC and applied to Zeta-Probe membranes using a Minifold I1 slot blot apparatus according to the manufacturer's directions (Schleicher and Schuell). Each slot blot also contained 2 pg and 4 p g of rabbit tRNA (BRL) to monitor for nonspecific hybridization. Each slot blot was done in duplicate. All hybridization experiments to the [32P]hsc70or the [32P]hsp70probes were done in one experiment. Hybndization and wash conditions for the [32P]hsp70,[32P]hsc70, and [32P]18SrRNA probes were done exactly as described for Northern blot hybridizations. The duplicate membranes probed with ["P]hsc70 oligonucleotide or [32P]hsp70oligonucleotide were stripped of any signal by repeated washings in 0.1 X SSC, 0.05% NaPPi, and 35% formamide at 100°C prior to hybridization with the [32P]18s rRNA oligonucleotide. The total period of time in stripping solution never exceeded 2.5 h. All pairs of duplicate blots were hybridized concurrently to the 18s rRNA oligonucleotide probe. Autoradiographic exposures, in the linear range for each probe, were scanned with a Kontes fiber optic scanner. Peaks were weighed, and each value for the hsp70 and hsc70 mRNAs was divided by its corresponding 18s rRNA value. The average value for the 2-pg and 4-pg signals was calculated for each sample and for each probe.

Temperature-versus-time profiles and total thermal response The body temperatures and time coordinates recorded for each animal were plotted on the same scale, and individual temperature-versus-time profiles were constructed. The average baseline temperature of control rats was 38.0"C. The area under each curve above the 38°C baseline was calculated. This value was termed the total thermal response (TTR) and is equivalent to the thermal response index used by Clark and Cumby (1975) and Glyn and Lipton (1980). It represented the magnitude and duration of each animal's hyperthermia in units of minute-degrees Celsius (min-"C) (Clark and Cumby, 1975). In the present study, one unit of TTR is equivalent to a 1"C change in body temperature lasting for 1 min.

Statistical analyses We performed an analysis of variance to compare mean hsc70 and hsp70 mRNA levels in cerebral cortex or cerebellum in control animals in relation to levels in amphetaminetreated or heat-shocked animals. A Student-Newman-Keuls test was used to find significant differences among groups using pairwise comparisons. A Student's t test was performed to compare the levels of hsc70 and hsp70 mRNAs in cortex versus cerebellum of control, heat-shocked, and amphetamine-treated rats. An analysis of variance was used to test the hypothesis that hsp70 and hsc70 mRNAs were induced in parallel with respect to TTR values. If slopes were equal, equality of intercepts on the y-axis were tested on lines generated by regression analysis. Nonparametric analyses were performed using Welch's t test and Mann-Whitney U test on small sample sizes (n < 5) (Zar, 1984).

2063

HEAT SHOCK PROTEIN 70 mRNA EXPRESSION IN RAT BRAIN

Oligonucleotide probes Specific oligonucleotide probes were synthesized to distinguish between the strictly inducible hsp70 mRNAs and the constitutively expressed hsc70 mRNAs. The sequences of the hsp70 and hsc70 oligonucleotide probes are shown in Tables I and 2. Each probe sequence is compared to the corresponding regions of other hsp70 and hsc70 coding sequences. Because no rat hsp70 genes have been cloned, the strategy for identifying a unique hsp70 sequence for oligonucleotide probe synthesis relied on comparative sequence data. By inference, a sequence that contained the fewest number of mismatches between phylogenetically distant species, such as humans and Drosophila, was very likely to be conserved in rat hsp70 gene sequences. Results of these comparisons are shown in Table 1. A 30-nucleotide region, with 50% GC base content, was identified. Nucleotides 369-398 of the human hsx70 sequence (Hunt and Morimoto, 1985) were conserved more between human hsx70, human hsp70, and Drosophila hsp70 than between rat hsc70, mouse hsc70, and human hsc70 sequences. The probe sequence contained only one terminal nucleotide mismatch to human hsp 70 and three nucleotide mismatches to Drosophila hsp70. By contrast, in the corresponding region of aligned hsc70 sequences, mouse hsc70 was mismatched by four nucleotides, human hsc70 by five nucleotides, and rat hsc70 by six nucleotides. The third nucleotide positions of hsc70 codons contain a higher percentage of the bases A and T than the third nucleotide positions of hsp70 codons. An oligonucleotide synthesized from the region of the rat hsc70 coding sequence corresponding to the region of the hsp70 probe would have a GC base composition of only 30%. Because conditions for hybridization stringency are related in part to GC content, it was necessary

RESULTS Body temperature-versus-time profiles for heatshocked and amphetamine-treated rats Figure 1A shows the mean body temperature-versustime profiles for the rats administered 10 mg/kg body weight dose of amphetamine sulfate (n = 12). As reported by Nowak (1988), this dose did not produce a uniform hyperthermic response. The 12 rats were divided into two groups based on the maximum body temperature that each animal reached. The mean values (_+SEM)are shown for each group. Group 1 amphetamine-treated rats (n = 7) were defined as those rats whose maximum body temperature was greater than 38°C and less than 40°C. These animals reached an average maximum body temperature of 39.1 f 0.6"C in 135 min. Group 2 amphetamine-treated rats (n = 5) were defined as those rats whose maximum body temperature was greater than 40°C. These animals reached an average maximum body temperature of 40.5 k 0.3"C in 120 min. As illustrated in Fig. lB, rats given the 15 mg/kg amphetamine dose (n = 12) showed a more consistent and rapid hyperthermic response. These rats reached an average maximum body temperature of 42.1 k 0.2"C in 70 min. Figure 1C shows the mean body temperature-versustime profiles for rats in the 90-min heat shock time course (n = 5). These rats reached an average maximum body temperature of 42.5 -t 0.3"C in 35 min. Figure 1D shows the mean body temperature-versustime profiles for rats in the 190-min heat shock time course (n = 4). These animals reached an average maximum body temperature of 42.0 k 0.1"C in 80 min. This heat shock time course was designed so that the body temperature-versus-time profiles paralleled the mean body temperature-versus-time profile of the 12 rats given the 15 mg/kg body weight dose of amphetamine.

42 41 40

FIG. 1. Mean temperature-versus-timeprofiles for rats made hyperthermic in the 10 mg/kg amphetamine study (A), the 15 mg/kg amphetamine study (B), the 90-min heat shock study (C), and the 190-min heat shock study (D). Error bars represent SEM. In A, A-----A depicts group 1 rats (body temperatures greater than 38"C,but less than 40°C), and 0 -0 depicts group 2 rats (body temperatures greater than 4OOC). Arrows denote time points for rats killed at early time points, and the numbers denote the number of rats killed at each time point. Single arrows represent group 1 rats, and double arrows represent group 2 rats.

39 38 37

w 38 a

T I . . ,

, , ,

,

.

I

.

.

m

.

.

,

1

, . I , . I

,+36 0

3 2 43

30

90 120 150 180 210 240

60

fD

oz n

w 42 w

41 40 39

38 37

-cI

6 36 3 0

30

80

90

120 150 180 210 240

0

' '

:

'

SO

'

I 60

"

:

SO

"

~

'

'

~

1 1I

"

' '

:

'

'

:-

120 150 180 210 240

TIME (MINUTES)

J. Neurochem.. Vol. 56, No. 6, 1991

E. K. MILLER ET AL.

2064

TABLE 1. Comparison oJhsc70 and hsp70 sequences to the hsp70 oligonucleotide probe” Nucleotide sequence no.

Aligned amino acid sequence no.

Gene

Species

hsx70 hsp70 hsp70 hsp68

Human Human Drosophila Mouse

369-398 360-389 1,312-1,341

121- I29 121-129 117-125

hsc70 hsc70 hsc70

Rat

41 1-440 425-454 1.721-1.750

121-1 29 12I- I29 121-1 29

Mouse Human

-

-

Nucleotide sequence

5’

3’

No. of bases mismatched to hsp70 probe

0

GTCCATGGTGCTGACCAAGATGAAGGAGAT

1

GTCCATGGTGCTGACCAAGATGAAGGAGAC

3 -

TTCGATGGTGCTGACCAAGATGAAGGAGAC

Area not sequenced

~TC~ATGGT~CTGAC&&ATGAAGGA&T CTCCATGGTTCTGAC&UGATGAAGG+T

6 4 5

TTCTATGGTTCTGACAAAGATGAAGGAAAT

The oligonucleotide probe sequence is written 3’ to 5’ and as the complement to the cDNA or gene sequence. The nucleotide sequence number and the aligned amino acid sequence number (Genbank notations) are shown for each sequence comparison. Sequence comparisons are written 5’ to 3’. The dots above nucleotides indicate nucleotides mismatched with the oligonucleotide sequence. The total number of mismatches are shown in the last column. References are as follows: human hsx70-Hunt and Morimoto, 1985;human hsp70-Voellmy et al., 1985;Drosophila hsp70-Ingolia et al., 1980;mouse hsp68-Lowe and Moran, 1986;rat hsc70-OMalley et al., 1985;mouse hsc70Giebel et al., 1987;human hsc70-Dworniczak and Mirault, 1987. a Sequence of the hsp70 probe: 3’ CAGGTACCACGACTGCTTCTACTTCCTCTA5‘.

to find a unique 30-nucleotide region in the rat hsc70 coding sequence with a 50% GC base composition that was equivalent to the 50%GC base composition of the hsp70 probe. As shown in Table 2, nucleotides 1,4131,442 (OMalley et al., 1985) were chosen for the hsc70specific probe. This region of the rat hsc70 sequence was mismatched by two nucleotides to human hsc70 sequences and by four nucleotides to mouse hsc70. By contrast, in the corresponding regions of hsp70 coding sequences, mouse hsp68 was mismatched by nine nucleotides, human hsp70 by 10 nucleotides, and Drosophila hsp70 by 14 nucleotides. Neither of these oligonucleotide probes was sufficiently similar to the corresponding regions in a fourth member of this gene family, the glucose-regulated protein 78 (Munro and Pelham, 1986), for cross-hybridization to occur (results not shown).

2. In amphetamine-treated rats, the hsp70 probe identified two mRNA species 3.53 kb and 3.05 kb in size. The hsp70 mRNAs were not detectable in control rats. These results are consistent with known patterns of hsp70 mRNA expression. The hsc70 probe identified a 2.55-kb mRNA species in both control and amphetamine-treated rats. The intensity of the hybridizatien signal was higher for amphetamine-treated rats than for control rats, suggesting that hsc70 mRNA levels were increased following amphetamine-induced hyperthermia. The hsp70 and hsc70 probes also identified mRNA species of identical molecular weights in the cerebellum of amphetamine-treated rats and in both brain regions of heat-shocked rats (results not shown). The northern blot analysis demonstrated that neither the hsp70 probe nor the hsc70 probe cross-hybridized with any other abundant mRNA species.

Northern blot analysis Autoradiographic exposures of northern blot hybridization of total RNA from cerebral cortex of control and amphetamine-treated rats hybridized to the [32P]hsp70 or the [32P]hsc70probes are shown in Fig.

Relative levels of hsp70 and hsc7Q mRNAs in cortex and cerebellum A comprehensive slot blot analysis was performed to determine more accurately the relative levels of hsp70 and hsc70 mRNAs in cerebral cortex and cer-

TABLE 2. Comparison of hsc70 and hsp70 sequences to the hsc70 oligonucleotide probe“

Gene

Species

Nucleotide sequence no.

Aligned amino acid sequence no.

5‘

Nucleotide sequence

3’

No. of bases mismatched to hsc70 probe

hsc70 hsc70 hsc70

Rat Human Mouse

1,413-1,442 3,573-3,602 1,427-1,456

455-463 455-463 455-463

CCTGCTTGGGAAGTTTGAGCTCACAGGCAT

0

CCTGCTTGGCAAGTTTGAACTCACAGGCAT

2 4

hsp68 hsx70 hsp 70 hsp70

Mouse Human Drosophila Human

699-728

455-463 455-463 453-46 I -

CCTGCT~GGG