Assessing mammillary body damage and possible interventions in a mouse model of hypoxic-ischemic encephalopathy

Abstract

Background: One of the most prevalent causes of neonatal mortality and neurological impairments is hypoxic-ischemic encephalopathy (HIE). In developed countries approximately 3 in 1000 neonates are affected, with an even higher prevalence in non-developed countries. HIE occurs when there is a reduced blood and oxygen flow to the brain. While the hippocampus is the most researched brain area related to this condition, mammillary bodies (MB) are receiving increasing attention in relation to neurological conditions and seem to also be related to HIE. Currently, the only available treatment is therapeutic hypothermia (TH) with a small therapeutic window and limited effectiveness. However, preclinical studies have shown that mesenchymal stem cell (MSC) treatment can alleviate some of the functional outcomes related to the hippocampus. This study aims to investigate the involvement of MBs in the pathogenesis of HIE and the kinetics of MB damage, and to investigate the effectiveness of TH and MSC treatment on MB damage as a result of HIE. Methods: On P9, C57BL/6 mice underwent an unilateral ligation of the carotid artery followed by 10% hypoxia to induce HI-injury. Depending on the research question, the pups received TH treatment (32°C for 3h), intranasal MSC treatment (0.5 x 106 cells), or no treatment. To investigate involvement of the MBs in HIE, brain tissue was stained for calcium-binding proteins calbindin (CB) and parvalbumin (PV), Neuron-specific nuclear binding protein (NeuN), and myelin basic protein (MBP) for the Mammillothalamic tract (MTT). To determine the kinetics of HI-injury, brain tissue was stained 3h, 1 day, 2 days, and 3 days post-HI. On hippocampal level, the tissue was stained with Microtubule-associated protein 2 (MAP2) for tissue loss. Further stainings in relation to the MBs included TUNEL for cell death, glial fibrillary acidic protein (GFAP) for astrocyte influx and Ionized calcium-binding adapter molecule (Iba1) for microglia activation. All stainings were done on the following timepoints post-HI: 3 hours, 1 day, 2 days, and 3 days. The effectiveness of TH was researched with the use of a GFAP staining and the same was done for the effectiveness of the MSC treatment in addition to a hematoxylin and eosin (HE) for tissue loss, MBP, and CB staining. Results: Following HI injury, the Ca2+ uptake is reduced in the MBs following HIE, but neuron count is not affected. Additionally, the MTT is reduced in size in both sides of the MB. MB injury occurs approximately at the same time as hippocampal injury, at day 1. TUNEL+ count shows a peak in cell death in the MBs at D1 as well, while Iba1+ microglia show activation at D2 and D3, followed by formation of a GFAP-scar is observed at D3 post-injury. TH did not reduce the size of the GFAP-scar at 10 days post-HI. Moreover, MSC treatment did not prevent tissue loss, as shown by HE, MTT size reduction, or a decrease in CB+ cell count, measured at 28 days post-HI. Surprisingly, we observed an increase in GFAP-scarring in the MBs of MSC-treated animals. Conclusions: This study shows that MBs are affected by HIE as shown by tissue loss and scarring. This is comparable to what is seen in the clinic. These results indicate that the model is suitable to further study MB treatment options. Furthermore, this study shows the injury happens simultaneously to hippocampal injury, suggesting that these injuries are not dependent on each other. Additionally, the injury follows the following cellular timeline: cell death, tissue loss, microglia activation, and GFAP-scarring. Lastly, both TH and MSC are not effective in preventing MB injury as a result of HIE, giving proof why TH is not effective on all patients and showing that further research is needed to assess the possible efficacy of MSC treatment on the MBs as it is proven to be effective on hippocampal damage.