Fifteen adult male C57BL/6 mice weighing 23 g to 25 g (7-8 W) were used in the present study. Mice were randomly assigned to 3 groups : 5 mice to the 30-min MCAO group, 5 mice to the 60-min MCAO group, and another 5 mice to the sham group. The Institutional Animal Care and Use Committee at the University of Korea reviewed and approved the animal protocols used in this work.

Focal cerebral ischemia in mice was induced by occluding the middle cerebral artery (MCA) using a modified intraluminal filament technique1). After a midline neck incision, the left common and external carotid arteries were isolated and ligated, and a microvascular clip was temporarily placed on the internal carotid artery. An 8-0 nylon monofilament coated with silicon resin was introduced through an incision in the common carotid artery and was advanced 9 mm distal to the carotid bifurcation for occlusion of the MCA. The nylon monofilament was retrieved after 30-min in the 30-min MCAO group and after 60-min in the 60-min MCAO group. Mice in the sham group received a midline neck incision as well as left common and external carotid artery-isolation and ligation, but no further manipulations. The size of the thread was matched to the body weight to assure reproducible vascular occlusion.

Mice were killed 24 h after the operation, and their brains were processed in 2 ways : both hemispheres were sectioned into 3 slices, each 1.5-mm thick, to measure the infarct volume, and brain tissues from bilateral hippocampi were promptly extracted for HSP-70 protein analyses.

All 15 mice were used for infarction volume measurements. Brains were removed and cut into 1.5-mm-thick axial slices. Slices were placed into a 2% solution of 2,3,5-triphenyltetrazolium chloride and warmed to room temperature for 20 min. Slices were then treated with paraformaldehyde (4%) and refrigerated at 4℃ for 48 h before scanning. The infarct area of each section was measured using National Institutes of Health (NIH) ImageJ software version 9.0 (Bethesda, MD, USA), and the total infarct volumes were calculated. Ipsilateral and contralateral hemisphere volumes were also determined to correct for edema, which can artificially affect the infarct volume19). The following formula was used to calculate the edema-corrected infarct volume : corrected infarct volume=contralateral non-infarction volume-ipsilateral non-infarct volume.

For western blot analyses, both hippocampi of all the mice from each of the experimental groups were obtained. Protein extraction, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and western blot were performed based on the manufacturer's instructions10,21). Equivalent amounts of protein were separated using 8% SDS-PAGE and transferred to PVDF membranes, followed by immunoblotting with the primary antibody. Primary antibodies used were as follows : anti-β-actin [F-1804 (Sigma, USA, 1 : 4000)] and HSP-70 [sc-32239 (Santa Cruz Biotechnology, Santa Cruz, CA, USA, 1 : 500)]. The intensities of the HSP-70 and β-actin protein bands were quantified from at least 5 independent experiments using NIH ImageJ software version 9.0 and normalized to β-actin.

For HSP-70 ELISA, hippocampi were snap-frozen in liquid nitrogen and weighed. Samples were homogenized for 15 s using a tissue homogenizer (Omni International, TH2000), followed by the addition of 50 µL of cold (4℃) 1×extraction buffer from the HSP-70 ELISA kit (Abcam, ab133060; Abcam, Cambridge, MA, USA). Following 30-min incubation on ice, samples were sonicated on level 7 with 12 pulses (Sonic Dismembrator 60; Fisher Scientific, Chicago, IL, USA) and centrifuged for 10 min at 21000×g in a 4℃ refrigerated microfuge. The supernatant was diluted 1 : 5 in sample diluent 2 (a component in HSP-70 ELISA kit); 100-µL duplicates were used in the ELISA, and the remaining supernatant was used for protein assays. Excess supernatant was stored at -70℃. The protocol provided with the HSP-70 ELISA kit was followed without any modifications for the quantification of HSP-70 protein levels. A µQuant (Bio-Tek Instruments, Inc., Winooski, UT, USA) was used to measure absorbance at 450 nm, and data were analyzed using Ascent for µQuant.

For RT-PCR, bilateral hippocampi from all subjects were extracted using the same method as described above. Total RNA was extracted using the Trizol reagent and reverse transcribed using a complementary DNA (cDNA) synthesis kit (Invitrogen, USA) according to the manufacturer's instructions. The fidelity and specificity of sense and antisense oligonucleotides were tested using BLAST (NIH, Bethesda, MD, USA). cDNA was synthesized using a DNA thermal cycler (Perkin Elmer) with the following parameters : 60-min at 45℃, followed by 5-min at 95℃. cDNA (2 µL) was amplified with Taq polymerase (Promega, USA) and 10 pmol each of sense and antisense primers as follows : GAPDH : 5'-TGT GTC CGT CGT GGA TCT GA-3', 5'-CCT GCT TCA CCA CCT TCT TGA-3'; HSP 70.1 : 5'-TGC TTG GGC ACC GAT TAC TGT CAA GG-3', 5'-GCA GCT AGA CTA TAT GTC TTC CCA GGC TAC TG-3'; HSPa1a : 5'-TCA ACG GGC GCG ACC TGA AC-3', 5'-GCC ACA TAT CTG TCT CCT AGC CAG C-3'; and HSPa1b : 5'-TCA ACG GGC GCG ACC TGA AC-3', 5'-TCG GTG CCC AAG CAG CTA TCA-3'. cDNA levels were normalized using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as reference. PCR products were electrophoresed through a 1% agarose gel, and bands were visualized using GelStar gel staining (FMC BioProducts, Rockland, ME, USA). Invariant GAPDH mRNA was used as an internal control.

Statistical analyses were performed using the Statistical Package for the Social Sciences version 18.0 for Windows (SPSS Inc., Chicago, IL, USA). Continuous and categorical variables were presented based on the average [± standard deviation (SD)] and the frequency (n and percentage), respectively. Kruskal-Wallis tests and the Mann-Whitney U-tests were used to compare the distributions of continuous variables among groups according to normality tests. Two-tailed hypothetical tests were used, and a value of p<0.05 was considered statistically significant in all analyses.

Cerebral infarctions were found in the hemisphere ipsilateral to the occlusion (left side) in animals that were exposed to transient ischemia (5 mice from the 30-min ischemia group and 5 mice from the 60-min ischemia group), whereas no focal infarction was noted in mice from the sham group (n=5) (Fig. 1A).

Based on the data from the mice in the 2 groups that had infarctions, cerebral infarction volume increased with MCA occlusion time. The average infarct volumes were 22.28±7.31 mm³ in the 30-min occlusion group (n=5) and 38.06±9.53 mm³ in the 60-min group (n=5), showing that the infarct volumes were greater in the group that experienced ischemia for a longer time. Comparison of infarct volumes showed a statistical difference between the 30- and 60-min MCAO groups (p=0.032) (Fig. 1B).

We processed bilateral brain samples from 15 mice for protein measurement (ipsilateral n=15, contralateral n=15). However, 1 sample from the ischemic side (left) in the 30-min ischemia group was unsuitable for analysis due to a flaw in the HSP-70 protein band. Therefore, 29 protein bands were analyzed : 14 ipsilateral protein bands and 15 contralateral protein bands.

Western blot analyses confirmed a significant increase in the ratio of HSP-70 to β-actin processed from ipsilateral (left) hippocampal tissues in both the infarction groups (30- and 60-min occlusion groups) compared to the sham group (sham group, n=5, 30-min group, n=4, 60-min group, n=5) (p=0.034).

However, there was no statistically significant difference in the level of HSP-70 between the 2 infarction groups (p=0.730). Using contralateral (right) hippocampal tissues, the ratio showed no significant change relative to different occlusion times (p=0.792). Finally, the expression of HSP-70 was greater in the tissue from an ipsilateral (left) side than that from the contralateral (right) when viewed in each subject (Fig. 2).

Similar to the western blot results, ELISA demonstrated an increase in the expression of HSP-70 in both infarction groups than in the sham group, when performed on hippocampal tissues ipsilateral to the occlusion (sham group : n=5, 30-min group : n=5, 60-min group : n=5) (p=0.011). However, there was no significant difference in HSP-70 expression level between the 30- and 60-min occlusion groups (p=0.971) despite increased infarction volumes with longer occlusion times. No statistical difference in HSP-70 expression between the groups was found when processing contralateral (right) hippocampal tissues, even for the group that underwent the longest occlusion time (p=0.058). Finally, HSP-70 expression was greater in tissue from the affected (left) side compared to the contralateral (right) side, when viewed in each animal from the 30- and 60-min occlusion groups (Fig. 3).

We measured the relative mRNA expression of GAPDH, hsp70.1, hspa1a, and hspa1 from hippocampal tissues obtained from both hemispheres (sham group : n=5, 30-min group : n=5, 60-min group : n=5). Quantitative analyses revealed no significant differences for the mRNA expression of the different HSP-70 subtypes (hsp70.1, hspa1a, hspa1b) between the groups that underwent different vascular occlusion times (sham, 30-min, and 60-min).

Upon comparing the data obtained from the infarcted hemispheres (ipsilateral to the vascular occlusion) using RT-PCR, we did not observe any significant differences in the expression of hsp70.1 (p=0.379), hspa1a (p=0.264), and hspa1b (p=0.33) mRNAs with regard to different occlusion times. We obtained the same negative results for contralateral (right) brain samples (hsp70.1, p=0.512; hspa1a, p=0.852; and hspa1b, p=0.932) (Fig. 4).