Combined effects of β-hydroxybutyrate and therapeutic hypothermia in a neonatal hypoxia-ischemia model

Abstract

Hypoxia-ischemia (HI) is one of the leading causes of brain damage during the development of newborns. It can result in death or cause varying degrees of neurological disability. The only well-established treatment currently available for neonatal HI is therapeutic hypothermia (TH). However, TH is only partially protective, reducing severe disability by approximately 11%. Therefore, new therapeutic approaches are urgently needed. It is known that immature brains utilize higher levels of ketone bodies, such as β-hydroxybutyrate (BHB), that may contribute to resistance to hypoxic-ischemic events. In this study, 11-day-old animals were subjected to the neonatal HI (Rice-Vannucci model) and treated with TH alone or in combination with BHB administration. To assess brain metabolism, glucose uptake was evaluated using MicroPET at 72 hours post-injury and when the animals reached 65 days of age. Behavioral tests, brain volume analysis, hippocampal cell counting and the assessment of hippocampal inflammatory cytokines expression were also performed. Animals treated with BHB exhibited increased glucose uptake at 72 hours post-injury and a reduction in neuronal loss in the hippocampus. The combined use of BHB and TH resulted in enhanced hippocampal neuronal survival, suggesting that BHB may represent a promising future treatment for neonatal HI.

Keywords: Hypoxia-ischemia; MicroPET; brain metabolism; therapeutic hypothermia; β-hydroxybutyrate.

Figure 1.

Experimental design. On the eleventh postnatal day (P11), animals were subjected to the HI procedure. Animals were divided into 5 groups (Sham, HI, HIB, HIT, HIBT); all animals except animals those in the Sham group, underwent the HI procedure. Animals in the HIB and HIBT groups received 4 β-Hydroxybutyrate (BHB) injections (at 0 h, 2 h, 4 h, and 6 h after hypoxia). HIT and HIBT groups were exposed to hypothermia (32°C for 5 h). On P14 and P65, the same pups from all groups were subjected to microPET imaging. Other animals from all groups were euthanized on P14 for cytokine expression analysis. On p65, the animals were submitted to the Morris water maze test, and after the test they were euthanized for histological analysis.

Figure 2.

Expression of cytokines in the hippocampus ipsilateral to carotid occlusion 72 h after the lesion. GM-CSF (a), IL-1α (b), IL-1β (c), IL-2 (d), IL-L6 (e), IFNγ (f), TNFα (g), IL-10 (h). Data were analyzed by one-way ANOVA followed by Tukey post hoc test. **Significant difference compared to the Sham group (p < 0.05). # Significant difference compared to the HI group (p < 0.05). Data are expressed as mean ± SD.

Figure 3.

Upper panel: Representative images showing SUV averages for neonates (a) and adults (b) for each group. After 3 days of injury (a, c, and d), hypermetabolism was observed in the hippocampus and cerebral cortex ipsilateral to the lesion in the groups that received BHB as treatment, while the other groups showed a metabolic pattern similar to those of animals not subjected to the HI. After the animals reached 65 days (b, e, and f), no significant differences were observed between the groups. The effects of glucose uptake were established in the hippocampus and cerebral cortex ipsilateral to the lesion in the adult phase. Data were analyzed by one-way ANOVA followed by Tukey post hoc test. # Significant difference compared to the HI group (p < 0.05). Data are expressed as mean ±SD.

Figure 4.

Effect of BHB and hypothermia treatment in neonates subjected to the HI model on spatial learning in the Morris Water Maze. Reference memory protocol during five days of training (a) was analyzed by repeated measures two-way ANOVA followed by Bonferroni’s post hoc. Area under the learning curve (b) during the five days of training, probe trial latency to reach the platform location (c) and time spent in the target quadrant (d) were evaluated using one-way ANOVA followed by Tukey post hoc test. #Significant difference compared to the HI group (p < 0.05). **Significant difference compared to the Sham group (p < 0.05). Data are expressed as mean ± SD.

Figure 5.

Upper panel: images of brain coronal sections stained with hematoxylin and eosin, representative of the brain volume in the experimental groups. Lower panel: Percentage of volume of the hippocampus (a) and cerebral cortex (b) ipsilateral to carotid occlusion in animals 2 months after lesion. Data were analyzed by one-way ANOVA followed by Tukey post hoc test. **significant difference compared to the Sham group (p < 0.05). Data are expressed as mean ± SD. Bar = 0.5 cm.

Figure 6.

Representative images of GFAP, Iba-1 and NeuN immunofluorescence for each experimental group in the CA1 area of the hippocampus (a) and in the cerebral cortex (e) ipsilateral to carotid occlusion. Count of positive cells for GFAP, Iba-1 and NeuN in the CA1 area (b, c, d) and cerebral cortex (f, g, h) ipsilateral to carotid occlusion in animals 2 months after lesion. Data were analyzed by one-way ANOVA followed by Tukey post hoc test. #Significant difference compared to the HI group (p < 0.05); **Significant difference compared to the Sham group (p < 0.05). Data are expressed as mean ± SD. Bar = 100 μm.