In total, 84 male specific pathogen-free grade Sprague Dawley rats (weight, 350–400 g; age, 8 weeks) were purchased from the Laboratory Animal Center of Guangzhou University of Chinese Medicine (Guangzhou, China). Firstly, 18 rats were randomly separated into three groups as follows: Group I, control group; Group II, the CA model group that received cardiopulmonary resuscitation (CPR) 4 min after CA; and Group III, the CA model group, which received CPR 6 min after CA. Secondly, 36 rats were then randomly separated into two groups, including 6 control rats and 30 CA model rats. CPR was conducted 6 min after CA. The CA rats were equally sub-divided into five groups, according to the different time points (6, 12, 24, 48 and 72 h) after ROSC. Thirdly, 30 rats were randomly divided into two groups, including six control rats and 24 CA model rats. These CA rats were equally sub-divided into SH group and controlled normothermia (CN) group. The SH and CN groups were again divided into two, determined by 24 h (SH Group I and CN Group I) or 48 h (SH Group II and CN Group II) after ROSC. The mortality rate of rats during the preparation of rat models was 6.7%. The present study was approved by the ethics committee of Sun Yat-Sen University (Guangzhou, China).

Rat CA and resuscitation were performed, as previously described (21,22). Anesthesia was achieved using 45 mg/kg pentobarbital sodium. Animals were ventilated with a Harvard Rodent Ventilator (Harvard Apparatus, Holliston, MA, USA), and the temperature was measured and maintained at 37±0.5°C throughout the preparation, insult, and first hour of recovery. Prior to asphyxia, the anesthetic gases were washed out with 3 min of ventilation at 100% oxygen, followed by 2 min of room air. Vecuronium was administered prior to asphyxia, in order to prevent reflex respiratory efforts during asphyxia. Subsequent to the washout period, animals were asphyxiated by disconnecting the respiratory tubing from the ventilator for 6 min, resulting in ~5 min of CA.

Following exactly 5 min, the ventilator was reconnected and ventilation was resumed with oxygen at a rate of 60 breaths/min. Intravenous epinephrine (0.005 mg/kg) and bicarbonate (1.0 mEq/kg) were administered, and external chest compressions were performed at a rate of 250–300 compressions/min. Rats generally experienced a ROSC within 2 min. If not, an additional dose of epinephrine was administered. Following stabilization for at least 60 min and confirmation of adequate spontaneous respirations, rats were extubated and weaned from oxygen back to room air.

In the CN group, rats were maintained at 37±0.5°C during CA and for 4 h after resuscitation. In the SH group, rats were maintained at 37°C during CA, but following resuscitation were allowed to regulate their own temperature.

General neurologic status was assessed using the validated NDS at 24, 48, and 72 h after CA (23). The score includes an assessment of consciousness, respiration, cranial nerve activity, motor and sensory function, and coordination. Normal rats have an NDS of zero.

Arterial blood samples (10 ml) were drawn at different time points after ROSC. Concentration changes in serum S100 calcium-binding protein B [S100B; cat. no. H6-KA0037, Multi Sciences (Lianke) Biotech Co., Ltd., Hangzhou, China], IL-1β [cat. no. 70-EK201B1/2, Multi Sciences (Lianke) Biotech Co., Ltd.] and IL-18 (cat. no. RK-KOA0362; Rockland Immunochemicals, Inc., Limerick, PA, USA) in the cerebral cortex were monitored by ELISA for the different groups. After coating with coating buffer (50 mmol/l sodium carbonate buffer, pH=9.6), plates were sequentially washed with phosphate buffered saline with 0.05% Tween-20 (PBST) buffer (cat. no. A100235-0001; Sangon Biotech Co., Ltd., Shanghai, China), blocked with 1% bovine serum albumin (BSA; cat. no. A602440-0050, Sangon Biotech Co., Ltd.), and incubated for 1 h at 37°C. Anti-S100B (1:200), anti-IL-1β (1:200) or anti-IL-18 antibodies (1:100), and horseradish peroxidase-conjugated antibody (1:1,000; all included in the corresponding ELISA kit) were sequentially added and incubated for 1 h at 37°C. The chromogenic substrate 3,3′,5,5′-Tetramethylbenzidine was added for detection. Absorbance was measured at a wavelength of 450 nm using an EnSpire multimode plate reader (Perkin Elmer, Waltham, Massachusetts).

Tissue samples were lysed in radioimmunoprecipitation buffer (50 mM Tris-HCl buffer pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% NP-40 and 0.25% sodium deoxycholate). Total protein was extracted from tissue lysate by centrifugation at 12,000 × g for 10 min at 4°C and protein concentrations were measured using a Bicinchoninic Acid assay (Sangon Biotech Co., Ltd.) according to the manufacturer's instructions. A total of 2 µg protein was loaded onto each lane of 10% polyacrylamide gel (250V voltage for 2 h) and blotted onto a polyvinylidene difluoride (PVDF) membrane. After blocking with PBST containing 5% nonfat dry milk, the membrane was incubated with antibodies against NLRP3 (cat. no. 13158; 1:500), apoptosis-associated speck-like protein containing a CARD (ASC; cat. no. 67824; 1:500), caspase-1 (cat. no. 2225; 1:400), caspase-3 (cat. no. 9662; 1:500), and β-tubulin (cat. no. 2146; 1:500; all Cell Signaling Technology, Inc., Danvers, MA, USA). Peroxidase-linked anti rabbit IgG (Thermo Fisher Scientific, Inc., Waltham, MA, USA) served as a secondary antibody. These proteins were visualized using an ECL western blotting detection kit (GE Healthcare, Chicago, IL, USA).

The rats were anesthetized with an overdose of 150 mg/kg pentobarbital and then sacrificed. The whole brain was immediately removed and frozen on dry ice. Then, the cerebral cortex site was separated from the whole brain. The sections of the cerebral cortex (~1.5 cm³) were sequentially washed in dimethylbenzene and ethanol, before being blocked in 3% H₂O₂. All nonspecific binding sites were blocked for 30 min in PBS with 5% BSA. The sections were then incubated at 4°C with anti-NLRP3, anti-ASC, anti-caspase-1, or anti-caspase-3 antibodies (Cell Signaling Technology, Inc.) overnight. Biotin-conjugated secondary antibody (cat. no. ab97044; Abcam, Cambridge, MA, USA) was applied to the slides and incubated for 1 h at room temperature. Finally, 3,3′-diaminobenzidine (DAB)/H₂O₂ was added to the surface of the slide to develop the color at room temperature for 10 min. The slides were visualized using a Nikon ECLIPSE 90i.

Sections were perfused with dimethylbenzene and sequentially washed with different concentrations of ethanol and blocked in 3% H₂O₂. These sections were incubated in 5% BSA for 30 min and the fragmented DNA was labeled with the TUNEL reaction solution at 37°C for 1 h. Converter-peroxidase was added to the sections at 37°C for 30 min before the TUNEL-positive nuclei were visualized by adding the DAB staining solution. All images were captured using a Nikon ECLIPSE 90i.

Data are presented as means ± standard error of the mean and analysis was performed with GraphPad Prism 6 software (GraphPad Software, Inc., La Jolla, CA, USA). The normalized intensity for western blotting bands was measured with Image J software version 1.45 (National Institutes of Health, Bethesda, MD, USA). Unpaired Student's t-tests and one-way ANOVA with a Tukey post hoc test were used to determine significant differences. P<0.05 was considered to indicate a statistically significant difference.

In order to elucidate the role of CA in neural defects, physiological parameters were compared between the control group and the two asphyxia groups, determining that there was no significant difference in physiological parameters among the three groups (Table I). Similarly, no apparent alteration in the majority of physiological parameters was observed between the two asphyxia groups following resuscitation. However, the duration of ROSC and the duration without blood flow in the asphyxia group that received CPR at 6 min after CA (Group III) was significantly longer than in the asphyxia group that received CPR at 4 min after CA (Group II; Table II). In addition, it was found that anal temperature was lower in the two asphyxia groups when compared with the control group. In addition, the rectal temperature in Group III was lower than that in Group II at five consecutive time points (from 10–120 min), but not at 180 min after ROSC (Fig. 1A).

Subsequently, it was determined whether the various neural defects were present in the asphyxia models. As shown in Fig. 1B, the NDS score was significantly lower in Group III than in Group II at 24, 48 and 72 h after CPR. Additionally, the S100B serum concentration of Group II and III was significantly greater than that in Group I at 72 h after CPR (Fig. 1C). The immunohistochemistry results indicated that cell morphology in the cerebral cortex was normal in the control (Group I; Fig. 1D). By contrast, morphological damage was observed in the cerebral cortex of Group II, including a vacuole and swelling in the cytoplasm (Fig. 1E). Further severe damage was observed in Group III, where necrosis was observed in the cells of cerebral cortex (Fig. 1F). These data demonstrated that increasing duration of CA increased the severity of neural damage in the cerebral cortex and enhanced the incidence of inflammation.

As the observed neural defects were associated with inflammation, whether the expression levels of certain key inflammatory factors were altered following resuscitation in the CA models were investigated. The potential changes in physiological parameters between the control and five asphyxia groups that were resuscitated at different time points after ROSC were detected. No apparent changes in physiological parameters were observed among the six groups, as presented in Table III. Furthermore, no significant difference in duration of asphyxia and resuscitation among the six groups was observed (Table IV). Whether the expression levels of key components of the inflammasome, including NLRP3, ASC and caspase-1 were altered following ROSC were determined. The expression level of NLRP3 was constantly demonstrated to increase at different time points, except for 48 h after ROSC (Fig. 2A and B). In addition, it was found that the expression level of ASC significantly increased at 6 h after ROSC, but did not exhibit any subsequent further increase (Fig. 2A and C). The expression level of caspase-1 increased at 6 and 48 h after ROSC, but showed no further increase at 12, 24 or 72 h after ROSC compared with the previous time points (Fig. 2A and D). These findings were confirmed by immunohistochemical analysis. The number of ASC, NLRP3 and caspase-1 positive cells significantly increased at different time points after ROSC, compared with what was observed in the control (Fig. 3A). Collectively, these results indicate that the expression levels of NLRP3, ASC and caspase-1 increase with increasing time following ROSC, further enhancing inflammation.

ELISA assays were performed to determine whether the concentrations of IL-1β and IL-18 in the cerebral cortex increased after ROSC. Compared with the control group, the concentration of IL-1β was significantly higher at 6 h and reached the maximal level at 12 h after ROSC (Fig. 3B). The concentration of IL-1β concentration gradually decreased at 24, 48 and 72 h after ROSC, but continued to be higher than that of the control group (Fig. 3B). Similarly, the IL-18 concentration was significantly higher after 6 h, and consistently increased with increasing time after ROSC (Fig. 3C). These data demonstrated that IL-1β and IL-18 concentrations were elevated in CA rat models after ROSC, indicating the aggravation of inflammation in the cerebral cortex.

Subsequently, the effects of SH on the neurological deficiency and inflammation observed in CA models were elucidated. No significant differences in physiological parameters were observed between the control, SH and CN groups (Table V). In the CA models, no apparent alteration in asphyxia time and CPR to ROSC time was identified, nor was any change observed in the tested physiological parameters among all four asphyxia groups following ROSC, as presented in Table VI. By contrast, the rectal temperature was significantly lower in the SH group than in the CN group (Fig. 4A). The NDS score was significantly higher in the SH group compared with the CN group at 24 and 48 h after ROSC (Fig. 4B). Additionally, the S100B concentration was shown to be reduced in the SH group compared with the CN group at 24 and 48 h after ROSC (Fig. 4C). Western blot analysis demonstrated that the expression levels of NLRP3, ASC, caspase-1 and caspase-3 markedly decreased in the SH groups compared with the CN group (Fig. 4D-H). Similar results were obtained via immunohistochemical analysis. The number of NLRP3, ASC, caspase-1 and caspase-3 positive cells in the SH group significantly decreased compared with the CN group (Fig. 5A). In addition, apoptosis was partially inhibited in the SH group, as the ratio of TUNEL-positive cells in this group was markedly reduced when compared with the CN group, while there was a limited number of apoptotic cells in the control group (Fig. 5B). Furthermore, SH after ROSC markedly decreased the concentrations of IL-1β and IL-18 (Fig. 5C). Finally, the cells in the cerebral cortex exhibited more severe damage in the CN group than did the cells in the SH group, while the cell morphology in the control group was normal (Fig. 5D). Thus, these data demonstrate that SH alleviated neural defects, apoptosis, and inflammation in the cerebral cortex, when compared with the CN group.