BackgroundTherapeutic hypothermia (TH) is the standard of care for neonates with hypoxic-ischemic encephalopathy, but it is not fully protective in the clinical setting. Hypoxia-ischemia (HI) may cause white matter injury (WMI), leading to neurological and cognitive dysfunction.MethodsP9 mice were subjected to HI as previously described. Pups underwent 3.5 h of systemic hypothermia or normothermia. Cresyl violet and Perl's iron staining for histopathological scoring of brain sections was completed blindly on all brains. Immunocytochemical (ICC) staining for myelin basic protein (MBP), microglia (Iba1), and astrocytes (glia fibrillary acidic protein (GFAP)) was performed on adjacent sections. Volumetric measurements of MBP coverage were used for quantitative analysis of white matter.ResultsTH provided neuroprotection by injury scoring for the entire group (n=44; P<0.0002). ICC analysis of a subset of brains showed that the lateral caudate was protected from WMI (P<0.05). Analysis revealed decreased GFAP and Iba1 staining in hippocampal regions, mostly CA2/CA3. GFAP and Iba1 directly correlated with injury scores of normothermic brains.ConclusionTH reduced injury, and qualitative data suggest that hippocampus and lateral caudate are protected from HI. Mildly injured brains may better show the benefits of TH. Overall, these data indicate regional differences in WMI susceptibility and inflammation in a P9 murine HI model.

The neonatal brain is highly susceptible to oxidative stress as developing endogenous antioxidant mechanisms are overwhelmed. In the neonate, superoxide dismutase (SOD) overexpression worsens hypoxic-ischemic injury due to H2O2 accumulation in the brain. Erythropoietin (EPO) is upregulated in 2 phases after HI, early (4 h) and late (7 days), and exogenous EPO has been effective in reducing the injury, possibly through reducing oxidative stress. We hypothesized that exogenous EPO would limit injury from excess H2O2 seen in SOD1-overexpressing mice, and thus enhance recovery after HI. We first wanted to confirm our previous findings in postnatal day 7 (P7) SOD-tg (CD1) mice using a P9 model of the Vannucci procedure of HI with SOD-tg mice from a different background strain (C57Bl/6), and then determine the efficacy of EPO treatment in this strain and their wild-type (WT) littermates. Thus, mice overexpressing copper/zinc SOD1 were subjected to HI, modified for the P9 mouse, and recombinant EPO (5 U/g) or vehicle (saline) was administered intraperitoneally 3 times: at 0 h, 24 h, and 5 days. Injury was assessed 7 days after HI. In addition, protein expression for EPO and EPO receptor was assessed in the cortex and hippocampus 24 h after HI. With the moderate insult, the SOD-tg mice had greater injury than the WT overall, confirming our previous results, as did the hippocampus and striatum when analyzed separately, but not the cortex or thalamus. EPO treatment worsened injury in SOD-tg overall and in the WT and SOD-tg hippocampus and striatum. With the more severe insult, all groups had greater injury than with the moderate insult, but differences between SOD-tg and WT were no longer observed and EPO treatment had no effect. Increased protein expression of EPO was observed in the cortex of SOD-tg mice given recombinant human EPO compared to SOD-tg given vehicle. This study confirms our previous results showing greater injury with SOD overexpression in the neonatal brain after HI at P7 in a different strain. These results also suggest that EPO treatment cannot ameliorate the damage seen in situations where there is excess H2O2 accumulation, and it may exacerbate injury in settings of extreme oxidative stress.

Background

Hyperpolarized 13C spectroscopic magnetic resonance spectroscopy (MRS) is an advanced imaging tool that may provide important real-time information about brain metabolism.

Methods

Mice underwent unilateral hypoxia-ischemia (HI) on postnatal day (P)10. Injured and sham mice were scanned at P10, P17, and P31. We used hyperpolarized 13C MRS to investigate the metabolic exchange of pyruvate to lactate in real time during brain development following HI. 13C-1-labeled pyruvate was hyperpolarized and injected into the tail vein through a tail-vein catheter. Chemical-shift imaging was performed to acquire spectral-spatial information of the metabolites in the brain. A voxel placed on each of the injured and contralateral hemispheres was chosen for comparison. The difference in pyruvate delivery and lactate to pyruvate ratio was calculated for each of the voxels at each time point. The normalized lactate level of the injured hemisphere was also calculated for each mouse at each of the scanning time points.

Results

There was a significant reduction in pyruvate delivery and a higher lactate to pyruvate ratio in the ipsilateral (HI) hemisphere at P10. The differences decreased at P17 and disappeared at P31. The normalized lactate level in the injured hemisphere increased from P10 to P31 in both sham and HI mice without brain injury.

Conclusion

We describe a method for detecting and monitoring the evolution of HI injury during brain maturation which could prove to be an excellent biomarker of injury.