Poster Z12 - Cortical spheroids as 3D in vitro models of traumatic brain injury-induced microglia activation and neuroinflammation

Introduction:

Introduction: Traumatic brain injury (TBI) is one of the leading causes of death and disability worldwide. Following a primary injury resulting from mechanical forces, resident immune cells of the brain called microglia become activated and trigger neuroinflammation. During neuroinflammation, microglia play an important role in restoring homeostasis by secreting pro- and anti-inflammatory cytokines while also undergoing morphological changes to migrate to sites of injury and clear cellular debris. Unresolved neuroinflammation, observed in moderate to severe TBI patients, can lead to the onset of neurodegeneration because pro-inflammatory microglia can induce secondary injury that triggers cellular death. The timing of microglia activation, the duration of microglia-mediated neuroinflammation and the dependence on injury severity, are not well understood. Here, we report the use of a 3D rat cortical spheroid model to study the effect of compressive strain in microglial morphology and motility, following acute (2-24 hours) and subacute (3-7 days) injury induced via centrifugation to understand the early onset of neuroinflammation in injured cortical microtissues. Cortical spheroids contain many desirable features of brain tissue, including: (a) being composed of multiple cell types such as neurons, astrocytes, microglia, and oligodendrocytes (Dingle 2015), (b) exhibiting electrical activity and neuronal networks (Dingle 2015, Sevetson 2021), (c) forming capillary-like networks and myelinated axons (Boutin 2018), and (d) being void of peripheral immune cells, which allows for the study of neuroinflammation mediated solely by microglia present in these spheroids.

Materials and

Methods: Materials and

Methods: Cortical spheroids were made by seeding multiple cells isolated from dissected neonatal (P2) rat cortices, in agarose microwells at an initial cell seeding density of 8,000 cells/spheroid (Dingle 2015). Spheroids were cultured for 14 days in vitro (DIV 14) and subjected to compression via centrifugation at angular velocities of 0 and 419 rad/s that resulted in compressive pressures of 0 and 149 kPa (González-Cruz 2023). To determine strain magnitude resulting from these pressures, finite element analysis was used to estimate compressive strains (0, and 35%, respectively). Control and injured live spheroids were labeled with isolectin IB4 (IB4), a dye that selectively labels microglia. We conducted 3D time lapse recordings of live microglia using a confocal microscope (Nikon AX) with a temperature/CO2-controlled stage. IB4-labeled microglia were imaged at 2 and 24 hours post-injury. Microglial morphological and motility changes were quantified using MotiQ plugin in Fiji software (Hansen 2022). Another subset of spheroids was fixed and immunolabeled with rabbit and mouse primary antibodies against ionizing calcium binding adaptor protein (Iba1, a pan microglial marker) and nuclear factor kappa-beta (NF-κβ) to examine NF-κβ nuclear translocation in microglia, a hallmark of pro-inflammatory gene activation. Iba-1 labeled microglial morphology was characterized via Strahler analysis using the SNT Neuroanatomy plugin in Fiji software (Arshadi 2021). 1-way ANOVA with a Tukey post hoc test were performed for statistical analysis using GraphPad Prism 9.0.

Results, Conclusions, and Discussions:

Results: Live microglia subjected to 35% strain exhibited a significant reduction in motility as early as 2h post-injury when compared to control microglia (p < 0.005). Microglial volume, branch number, and ramification index decreased in injured spheroids compared to controls (p < 0.005). Microglial branch extension and retraction rates and volume spanned by branch extensions and retractions also decreased in microglia from injured spheroids (p < 0.005). Dual Iba1 and NF-κβ immunostaining revealed few NF-κβ nuclear translocation events in round microglia at 24h post-injury but not in control microglia. Iba1 immunostaining also showed that compared to uninjured spheroids, , microglia in injured spheroids exhibited lower numbers of branches 3 days post-injury. By 7 days post-injury, microglia in injured spheroids showed higher numbers of branches and branch lengths (p < 0.005).
Discussions: Microglia morphology underwent drastic changes in both motility and morphology, with changes in motility preceding morphological ones, suggesting that microglia became slower as a response to acute cell death following mechanical injury. These changes in morphology and motility suggest changes from a branched, surveillant phenotype to a round, ameboid shape that is often observed in pro-inflammatory microglia following acute TBI. Of note, the emergence of hyper-ramified microglia at day 7 post-injury resembled microglial phenotypes observed in chronic inflammation in animal TBI models. While NF-κβ nuclear translocation was only observed in round microglia, this event suggests that a subpopulation of microglia adopted a pro-inflammatory phenotype as early as 24h post-injury.

Conclusions: We have engineered a 3D in vitro, multicellular cortical spheroid model to study acute and subacute TBI-induced neuroinflammation. This model allowed us to study microglia in a controlled microenvironment that mimics key in vivo features of neuroinflammation. Understanding the complexity of microglial dynamics during the onset of neuroinflammation is critical for designing therapeutic strategies to resolve neuroinflammation and mitigate unwanted secondary injury.

References: Dingle et al, Tissue Eng Part C Methods, 21(12), 2015; Sevetson et al, Lab Chip, 21(23), 2021; Boutin et al, J Neurosci Methods, 299, 2018; González-Cruz et al, bioRxiv, 11.15.567286, 2023; Hansen et al, Mol Bio Cell, 33(11), 2022; Arshadi et al, Nat Methods, 18, 2021

Acknowledgements (Optional): We thank Geoffrey Williams for assistance with confocal microscopy and Alexander Del Toro for help with MotiQ microglial analysis. This research based upon research supported by the U. S. Office of Naval Research under PANTHER award number N000142112044 through Dr. Timothy Bentley. Also, RDGC receives additional funding support for research expenses from the Brown University's Carney Institute for Brain Science's Advancing Research Careers (ARC) program (R25NS124530).