(131d) An Organotypic Whole Hemisphere Slice Model of Mitochondrial Abnormalities in the Neonatal Brain

In the brain, mitochondria play critical roles in processes that dictate cell function, communication, and fate. The repertoire of roles mitochondria has in neural cells links metabolic dysfunctions to a variety of neurologic conditions, including stress, aging, neurodegeneration, traumatic brain injury, post-traumatic stress disorder, stroke, and neuropsychiatric disorders. Accumulating evidence suggests that mitochondria have important roles during neurodevelopment, but the systemic effects of mitochondrial abnormalities in the developing brain are less explored. The developing brain is especially sensitive to metabolic disruptions and there are several unique avenues by which neonatal populations experience stress and injury¹, highlighting the need to be able to study mitochondrial abnormalities in a dynamic system more representative of the neonatal brain environment.

Multiple studies capture mitochondrial abnormalities as outcomes of specific developmental injury and disease states and tend to focus on a specific region or cell type. Here, we leverage an organotypic whole-hemisphere (OWH) brain slice platform to model mitochondria-driven metabolic dysfunction in the neonatal brain with regional specificity and access to native cytoarchitecture. We have demonstrated that OWH slices obtained from postnatal (P) day 10 rats, term equivalent to the human neonate, retain viability and metabolic activity up to two weeks in vitro². This postnatal age and culturing window are well-suited to study healthy and abnormal mitochondrial dynamics in the neonatal brain. To systemically drive mitochondrial abnormality, we expose OWH slices to low-grade doses of mitochondrial toxin ROT (ROT), a potent inhibitor of respiratory chain protein Complex I. We examine the tissue-level, extracellular, cellular, and organelle responses to repeat ROT exposure and single ROT exposure in two distinct brain regions over 6 days in culture.

Methods

OWH slices are collected from P10 healthy female Sprague-Dawley rats, with methods described previously². Individual slices are plated onto 30-mm 0.4-μm-pore-sized cell culture inserts containing 1mL 37°C 5% slice culture media (SCM). At 4DIV, ROT is introduced to OWH slices via culture media. ROT solution aliquots (1 mM) were serially diluted into 37°C SCM to achieve the desired ROT concentration. At 4DIV, SCM is replaced with ROT media (RSCM) at 50 nM. For the repeat-hit exposure, slices are incubated with RSCM for 2 days and replaced with RSCM at 2-day increments out to 10DIV. For the single-hit exposure, slices are incubated with RSCM for 2 days and replaced with healthy media at 2-day increments out to 10DIV. Our healthy control slices are slices cultured without ROT, with media changes at identical time points. To screen whole-slice cytotoxicity in response to ROT exposure, culture supernatants were collected acutely, at 4DIV, then every 2 days out to 10DIV, and assayed for lactate dehydrogenasel². At 4DIV, slice media was replaced with 1 nM, 50 nM, 100 nM, or 10 μΜ RSCM. At multiple time points, OWH slices were characterized at the organelle, cellular, and extracellular levels following 50-nM or 10-μΜ exposure. For qualitative mitochondria studies, images of mitochondria were collected in live OWH slices using MitoTracker Deep Red FM³. At the same time points, separate OWH slices were assessed for density, % cell damage, and co-localization of both damage and density with mature neurons (anti-NeuN) and microglia (anti-Iba1), in the cortex and striatum. Cell death imaging and analysis of propidium iodide (PI) positive cells and co-staining for microglia and mature neurons was performed². For cell type analysis, Iba1+/NeuN+ cells were also counted manually after applying Otsu threshold and determining overlap with DAPI+ nuclei, then applied a 50-pixel size thresholds before manual counting. Image acquisition settings and fluorescent cutoffs were consistent for all images images. For MPT experiments in live brain slices, particles were tracked in ROT-exposed slices immediately after each exposure window, with no other changes to protocol and analysis².

Results

To first determine if we could sustain ROT-driven injury in our OWH slices over longer times without compromising overall OWH slice health, we measured overall OWH slice healthy via LDH release over 10 days (Fig 1A). LDH concentration increased significantly for all ROT concentrations higher than 1 nM, with 10 μM exposure causing the most injury. Two repeat exposures to 50 nM ROT consistently resulted in injury through 10DIV and relative injury from the 50 nM exposure level had minimal temporal dependence. The findings suggested that 50 nM ROT is most suitable for inducing consistent injury in OWH slices through 10DIV. A 10 μM exposure was also chosen as a positive control for ROT-induced injury in subsequent cytotoxicity studies. Regional cytotoxicity response to ROT exposure was also quantified (Fig 1C). In the cortex and striatum, median % cell damaged increased through 10DIV to reach 60% in the cortex and 42% in the striatum for a repeat exposure. Following a single exposure, the striatum maintained higher % cell damage compared to the cortex at 10DIV. We noted apparent inflection behavior at 8DIV for cell density and % cell damage by region. 50 nM ROT-exposed slices showed altered mitochondrial morphology at 6DIV and 8DIV (Fig 1B). Differences in mitochondrial morphology between 50 nM and 10 μM ROT were more distinct by 8DIV. Single-exposure slices recovered by 10DIV. Healthy slices displayed bright, punctated mitochondria localized to nuclei through 10DIV. These findings demonstrate ROT-induced mitochondrial abnormalities and their temporal and exposure-dependent evolution in OWH slices.

At the cellular level, following repeat exposure, cortical and striatal neurons showed comparable distribution and co-localization with PI-positive signal which increased through 10DIV (Fig 1D, F). For a single ROT exposure, cortical neurons appeared more resilient compared to striatal neurons. The regional differences may be partially explained by microglial interactions. ROT exposure increased microglial density increased in both regions, but more so in the striatum. Peak microglial density and PI-colocalization occurred at 8DIV, with sustained higher density in the cortex compared to the striatum by 10DIV (Fig 1E, G). For microglia in both regions, ROT exposure led to a shift in morphology towards a more pro-inflammatory state, which was sustained for a repeat exposure, suggesting an ongoing, evolving microglial response to neuronal injury. These findings suggested region- and time-dependent interactions between neurons and microglia in response to ROT-induced injury. At the extracellular level, single-hit ROT exposure initially slowed diffusion at 6DIV in the cortex, while repeat-hit ROT further reduced diffusion at 8DIV (Fig 1G). Unexpectedly, single-hit ROT led to faster diffusion at 8DIV, before reverting to slower diffusion by 10DIV. Conversely, in the striatum, both single and repeat exposures initially accelerated diffusion at 6DIV. MPT data highlighted dynamic relationships between microstructure and cellular interactions, demonstrating that changes in diffusivity were attributed to complex interactions between neural cells and the microenvironment, and not attributed solely to changes in cell density.

Conclusions

In this work, we investigated the effects of ROT exposure in OWH slices as a model of mitochondrial abnormalities in the neonatal brain. Our results highlight severity-, time-, and region-, and cell population specific responses. Live-slice imaging showed qualitative differences in mitochondrial fluorescence signal and morphology dependent on exposure time and severity. Regional cytotoxicity data showed increased cell damage over time, particularly in the cortex, suggesting regional susceptibility which was further explored through neuronal and microglial responses. Cytotoxicity trends were consistent with the observed responses in neuron and microglia populations, where ROT exposure influenced neuronal sensitivity and microglial activation differentially by region and culture time. The interplay between these cell types may contribute to the differential susceptibility of cortical and striatal regions to ROT-induced damage, which was also detected through MPT experiments. Ongoing work incorporates RT-qPCR screening of mitochondrial markers and complex I quantification via NAD+/NADH ratio assay. This model provides a platform for comparing adverse mitochondrial-driven outcomes in other developmental brain injury models and enable screening of therapeutics that target mitochondrial regulation.

References:

1. Zhao, T. et al. (2022). 10.1159/000526491
2. McKenna, M. et al. (2022). 10.1186/s13036-022-00293-w
3. Liao, R., et al. (2020). 10.1186/s13036-020-0226-8