TOP 74 - Human Milk Extracellular Vesicles for Neuroprotection in a Ferret Brain Slice Model of Preterm Brain Injury

Publication Number: 3806.TOP 74

Rachel laFontaine, University of Washington School of Medicine, Seattle, WA, United States; Kate F. DiNucci, University of Washington School of Medicine, Seattle, WA, United States; Zheyu Ruby Jin, University of Washington, Seattle, WA, United States; Colin J. Landis, University of Washington, Seattle, WA, United States; Elizabeth Nance, University of Washington, Seattle, WA, United States; Thomas Wood, University of Washington School of Medicine, Seattle, WA, United States; Rebecca Hoban, Seattle Children's, Seattle, WA, United States

Background: Preterm birth, a leading cause of neonatal morbidity and mortality worldwide, disrupts the most rapid period of in-utero brain growth. As a result, nearly half of all extremely preterm (EP) infants (born < 28 weeks' gestation) suffer from preterm brain injury (PBI). There is currently no standardized treatment to address the short- and long-term sequelae of PBI. Extracellular vesicles (EVs) isolated from human milk (HM) are easily accessible, cross the blood-brain barrier, and have immunomodulatory properties that are promising as a neurotherapeutic. Prior studies suggest beneficial effects of EVs and HM in neonatal models; however, to our knowledge, HM-derived EVs have yet to be investigated as a therapy for PBI.

Objective: To evaluate the effects of HM-derived EVs using an EP ferret organotypic whole hemisphere brain slice model of oxygen glucose deprivation (OGD), an ex vivo model of PBI.

Design/Methods: Our ex vivo slice model replicates important features of human neuroanatomy such as gyrification and white matter structure (Figure 1) while maintaining 3D neural architecture, allowing for region-specific analysis of injury and treatment responses. Brains of postnatal day (P) 14 ferrets (equivalent to EP infants) will be extracted, with 300µm slices obtained and cultured in glucose-enriched serum for 72 hours. Slices will then be randomized by sex to 3 groups (36 slices/group): control, OGD, and OGD + HM EVs. OGD-exposed slices will be placed in glucose-deprived media and into a hypoxic chamber for 3 hours (Figure 2), which models hypoxic-ischemic conditions that trigger PBI. As a pilot study, we will utilize HM samples from an individual consenting, anonymous donor. HM EVs will be isolated via cushioned ultracentrifugation, filtered through a 0.22μm syringe filter, and then concentrated by centrifugation through a 100kD cellulose filter. EV concentration will be measured using Nanosight Tracking Analysis. EV dosages of 1-5 x1011/kg will be applied to slices immediately following OGD, with harvesting after 24 h. Cell injury will be assessed using a lactate dehydrogenase (LDH) assay to quantify release of intracellular LDH, and nuclear staining with DAPI to quantify pyknotic nuclei via confocal microscopy imaging. Staining will be assessed globally and regionally within the basal ganglia, hippocampus, corpus callosum, cortex, thalamus, and subcortical white matter. We hypothesize that HM EVs will ameliorate damage following OGD, reducing LDH release and decreasing the number of pyknotic nuclei.

Figure 1
Comparison of different term-equivalent animal brains versus humans. Ferret brains are more comparable to larger animal models and human brains due to their white matter percentage and gyrification. This increases translatability for future tests with larger animals and clinical trials. Coronal slice of non-human primate brain used as proxy to display approximate human grey and white matter ratios.

Figure 2
Organotypic whole hemisphere oxygen-glucose deprivation model procedure. Beginning with slice collection from the P14 ferret which are plated in media for a 72 hour period before beginning OGD injury and 3 hour incubation. Finally, slices receive randomized treatment followed by a 24 hour incubation period before media collection and analysis. Figure courtesy of Ruby Jin, Nance Lab.