In both the normothermia and hypothermia groups, the body temperature of the pups followed very closely the ambient temperature in the chamber (Fig. 1). The rectal temperature of the pups remained fairly constant and stayed at about 35.5 °C in the normothermia group and at about 32 °C in the hypothermia group. There was no significant difference in rectal temperature at any time point, in either group, between control animals and animals subjected to HI (Fig. 1).#

To assess IL-18 mRNA expression after post-ischemic hypothermia in the immature brain, the level of IL-18 mRNA was investigated by real-time RT-PCR at 24 h and 72 h (Fig. 2). The IL-18 mRNA levels in the ipsilateral hemispheres of the normothermia group were significantly increased at 24 h (P=0.009) and 72 h (P=0.03), reaching a 2.1- and 2.1-fold increase compared with the control levels at P8 and P10, respectively. In contrast, in the ipsilateral hemisphere of the hypothermia group, the level of the IL-18 mRNA was significantly increased at 24 h after HI (P=0.002), but no significant difference was observed at 72 h after HI compared with the age-matched control. The IL-18 mRNA level in the ipsilateral hemisphere of the hypothermia group was significantly lower than the level of the normothermia group at 24 h (P=0.02) and 72 h (P=0.01) after HI.#

To assess the effects of hypothermia on the protein level of IL-18 in the immature brain, we investigated the protein level of IL-18 in the ipsilateral hemispheres of both normo- and hypothermia groups at 24 h and 72 h after HI by using Western blotting. The antibody detected a protein with an apparent molecular mass of 24 kDa, which corresponds to the size of pro-IL-18 (Fig. 3), but detected no cleavage (Hedtjarn et al., 2002). As there were no significant differences between normo- and hypothermia control groups, they were pooled as a control. At 24 h after HI, the protein levels were slightly increased in both the normo- and hypothermia groups, compared with the control animals, but no significant difference was detected by densitometric analysis. At 72 h after HI, the protein level in the normothermia group was significantly increased (P=0.02), but no significant increase was noted in the hypothermia group, compared with control animals. In addition, the IL-18 protein level in the ipsilateral hemisphere of the hypothermia group was significantly lower than the level in the normothermia group at 72 h (P=0.02) after HI.#

To examine the localization of and effects of HI on IL-18, histological sections at 72 h after HI were immunohistochemically stained with antibodies against IL-18. The uninjured control brains of both normo- and hypothermia groups were pooled as a control because there were no significant differences between them. In all groups, very few positive cells were detected in the contralateral hemispheres (data not shown) and age-matched uninjured control brains at this time point (Figs. 4(A) a, b). In the normothermia group at 72 h after HI, IL-18-positive cells were detected throughout the entire cortex, corpus callosum and striatum of the ipsilateral hemisphere (Figs. 4(A) c), and the positive cells showed a macrophage-like morphology with a relatively large round cell body and short processes (Figs. 4(A) d). In contrast, in the hypothermia group at 72 h after HI, very few positive cells were detected in the entire ipsilateral hemisphere (Figs. 4(A) e and f). The number of IL-18-positive cells was counted in the cortex and throughout the corpus callosum (CC) of the ipsilateral hemispheres of both the normo- and hypothermia groups and the uninjured brains of normo- and hypothermia groups (Figs. 4(B), (C)). The control brains of both groups were pooled as a control because of no significant differences between them. Positive cells were not detected in the control animals, and there were very few positive cells in the ipsilateral hemisphere of the hypothermia group. The numbers of IL-18-positive cells in both the cortex and CC were significantly increased at 72 h after HI in the ipsilateral hemisphere of the normothermia group compared with control animals (p<0.001) and the ipsilateral hemisphere of the hypothermia group (p<0.001).#

To clarify the cellular source of IL-18 post-HI in the immature brain, immunofluorescence double-staining was performed for IL-18 plus isolectin as a marker of microglia. There were numerous IL-18-positive cells in the ipsilateral hemisphere in the normothermia group, most of which were colocalized with isolectin-positive cells (Fig. 5), and morphologically identified as ameboid microglia, as judged by the criteria described in Experimental procedures. On the other hand, only a few double-stained cells were observed in the ipsilateral hemisphere in the hypothermia group. Instead, there were many cells positive for isolectin alone, with a small oval cell body with long branched processes in the ipsilateral hemisphere in the hypothermia group (data not shown).#

To investigate in detail the effect of temperature on the morphology of microglia after HI, the isolectin-positive cells were visualized using 0.5 mg/ml 3,3′-diaminobenzidine and classified into three types and quantified. In the normothermia group, the cortex, CC and striatum were severely damaged, as judged by the loss of MAP-2, where most of the isolectin-positive cells appeared to be ameboid microglia (AM) (Figs. 6e, f, g, h). However, few ramified microglia (RM) or primitive ramified microglia (PRM) were detected in the damaged area of the normothermia group. Microglia were also observed in the non-infarcted areas, but the number of positive cells was fewer and its morphologic modification was quite limited, which is consistent with a previous study (Sizonenko et al., 2005). In contrast, in the ipsilateral hemisphere of the hypothermia group, the loss of MAP-2 area was quite limited (Fig. 6i), which agrees with our previous studies using the same model (Fukuda et al. 2001; Tomimatsu et al. 2002 2003), and most of the isolectin-positive cells were either ramified microglia (RM) or primitive ramified microglia (PRM), although quite a few ameboid microglia (AM) (Figs. 6i, j, k, l) were also observed.#

Fig. 7 shows the number of each of the three morphologically classified types of microglia in the cortex (a) and CC (b) in the normo- and hypothermia group and control animals. The number of AM in the normothermia group was significantly increased in both the cortex (p<0.001) and CC (p<0.001) compared with the number in the control brains, but few RM or PRM were observed in these areas, suggesting that there was a marked increase in the number and activation status of microglia in the normothermia group. In contrast, the number of AM in the hypothermia group was not significantly increased in cortex and CC compared with the number in the control brains, and there were no significant differences in the number of RM+PRM between the ipsilateral hemispheres of the hypothermia group and control brains.#

Neonatal HI was induced in rats at postnatal day 7 (P7) according to the method described by Rice et al. (1981). Seven-day-old pups of either sex were anesthetized with halothane (4.0% for induction, 2.0% for maintenance) in room air, and the left common carotid artery was isolated and ligated with 5-0 surgical silk. The procedure was completed within 10 min. After the procedure, the pups recovered for 1 h in a temperature-controlled incubator. The pups placed in a chamber were perfused with a humidified gas mixture (8% oxygen in nitrogen) for 60 min. The temperature in the incubator, and the temperature of the water used to humidify the gas mixture, were kept at 36 °C.#

The pups were randomized into two groups at the end of hypoxic exposure. The normo- and hypothermia groups were placed in a chamber submerged in a water bath at a stable temperature of 36 °C or 32 °C, respectively, for 24 h without the dam. Control animals were anesthetized but not subjected to HI. After the 24-hour normo- or hypothermia period, all pups were returned to their biological dams until sacrificed. The rectal temperature was measured in all rats at 1 h, 6 h, 12 h and 24 h after HI using a rectal probe. It has been shown that the rectal temperature corresponds very well to the brain core temperature (Thoresen et al., 1996).#

Real-time RT-PCR was performed with a Gene Amp PCR System 2700 (Applied Biosystems; Lincoln, CA, USA). Animals were killed by decapitation at 24 h or 72 h after HI, and the brains were rapidly removed and the ipsilateral hemispheres were frozen in liquid nitrogen (n=6 animals at each time point per group). Total RNA was extracted from the hemispheres individually according to the manufacturer's instructions, using an Rneasy ® mini kit (Qiagen Inc.). First-strand cDNA synthesis was performed with a Superscript RNase H⁻ reverse transcriptase kit (Life Technologies, Inc.), oligo dT primers and dNTP (dATP, dCTP, dGTP and dTTP, Roche Molecular Biochemicals).#

Each PCR reaction mixture (20 μl) contained 1/20 of the cDNA synthesis reaction, 2 or 3 μM MgCl2 depending on the optimal concentration for each primer pair, 0.5 μM forward and reverse primers and 2 μl of LightCycle-FastStart DNA Master SYBR Green 1 (Roche), which contains FastStart Taq DNA polymerase, reaction buffer, dNTP mix and SYBR Green 1 dye. The following primer pairs (from CyberGene AB, Huddinge, Sweden) annealing temperature and elongation time were used:

• IL-18:forward: 5′-AAACCCGCCTGTGTTCGA-3′; reverse: 5′-TCAGTCTGGTCTGGGATTCGT-3′, 57 °C, 12 s.
• GAPDH:forward: 5′-TGCCAAGTATGATGACATCAAGAAG-3′; reverse: 5′-AGCCCAGGATGCCCTTTAGT-3′, 58 °C, 15 s.

The samples were initially warmed to 50 °C for 2 min, then the AmpliTaq Gold DNA polymerase was activated by heating to 95 °C for 10 min, annealing temperature (see above) for 4 s and 72 °C for 12–15 s (see above) depending on the product size. Each sample was assayed in duplicate. After the final cycle, the melting curve was determined to check that only one product had been produced, and the PCR product was electrophoresed on a 1.5% agarose/0.5 times TBE gel containing ethidium bromide to confirm that the product was of the expected size. For quantification and for estimating the amplification efficiency, a standard curve for each gene product was constructed using increasing concentrations of cDNA. The amplified transcripts were quantified by comparison with a standard curve, and the relative efficiency of expression was compared to that of GAPDH.#

Animals were killed by decapitation 24 h or 72 h after HI (n=4 animals per group). Control animals were killed on postnatal day 8 or 10. The brains were rapidly removed, and the ipsilateral hemispheres and control ones were frozen in liquid nitrogen. Each sample was homogenized in a lysis buffer (0.01 M Tris, pH 7.8, 0.1 M NaCl, 1 mM PMSF, pepstatin 2 μg/ml, leupeptin and chymostatin 2 μg/ml) and centrifuged at 12,000×g for 10 min at 4 °C. The protein concentration of the lysates was then determined with a BCA protein assay kit (Pierce, Rockford, IL, USA). Samples were denatured in a gel-loading buffer at 100 °C for 3 min and loaded on a 15% sodium docecyl sulfate-polyacrylamide gel. After electrophoresis, proteins were transferred to a nitrocellulose membrane. The membrane with the transferred proteins was incubated with goat anti-rat IL-18 antibody (0.12 μg/ml in PBS-T, AF521, R&D Systems, Minneapolis, MN) overnight at 4 °C. After washing, the membranes were incubated with the appropriate peroxidase-conjugated secondary antibodies diluted in blocking buffer. The blot was visualized using an enhanced chemiluminescence (ECL) Western blotting detection kit (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) and exposed to X-ray film. The relative amount of proteins was calculated from densitometric scanning.#

Another set of animals was treated with the combination of 60-min HI and 24-h thermal regulation procedure, as described above (n=5 animals per group). The animals were deeply anesthetized 72 h after HI and transcardially perfusion-fixed with 4% paraformaldehyde in 0.1 M PBS. The brains were removed and immersion-fixed in the same solution at 4 °C for 24 h, dehydrated with a graded series of ethanol and xylene, embedded in paraffin and cut into 5-μm coronal sections. Adjacent sections were immunostained using the following primary antibodies and working concentrations: IL-18 (10 μg/ml in PBS, R&D Systems, Minneapolis, MN), anti-MAP-2 (1:2000 in PBS; clone HM-2, Sigma). After deparaffinization and rehydration, antigen retrieval was performed by heating the sections for 10 min in 10 mM citrate buffer (pH 6.0). Nonspecific binding was blocked with 4% horse serum (depending on the species used to raise the secondary antibody) in PBS. After blocking, sections were incubated with primary antibodies at room temperature for 1 h followed by the appropriate biotinylated secondary antibodies for 1 h (Vector Laboratories, Burlingame, CA). Endogenous peroxidase activity was blocked with 0.3% H2O2 in methanol for 10 min, and visualization was performed using an avidin–biotin–peroxidase solution (Vectastain ABC Elite kit, Vector Laboratories) with 0.5 mg/ml 3,3′-diaminobenzidine. For double labeling immunohistochemistry, secondary antibodies directly conjugated to Texas Red or FITC were used. Microglia were detected using biotinylated isolectin B4 (10 μg/ml in PBS, 1 h, L2140, Sigma) or FITC-labeled isolectin B4 (L2895, Sigma).#

All values are expressed as mean±SD. ANOVA with Fischer's PLSD post hoc test was used to compare ipsilateral hemispheres from the hypothermia group with ipsilateral hemispheres of the brains of the normothermia group or control animals, and p<0.05 was considered statistically significant. Student's unpaired t-test was used to compare values between the hypo- or normothermia groups and control animals, and p<0.05 was considered statistically significant.#