Session: 05-01-02: Injury and Damage Biomechanics: Brain Response at the Cellular Level

Paper Number: 95179

95179 - Determination of Critical Cellular Traumatic Brain Injury Thresholds via in Vitro Mechanical Impacts and Deep Learning-Based Cell Phenotype Identification 

 

Mild traumatic brain injury (mTBI) caused by inertial, blunt or blast forces poses a significant threat to both civilian and military communities. Yet, the exact force magnitudes and directions that initiate TBI at the cellular level, i.e., critical cellular thresholds of injury, are still unknown. This information is critically important since the clinical effects of TBI are rooted, at least initially, in cellular signaling cascades arising from the mechanical insult to the head. With no established thresholds for these injuries, personal protective equipment and safety guidelines cannot be developed to provide adequate protection. Considering both the structural and cellular diversity of the brain, we developed a unique and extensive experimental in vitro approach for resolving phenotype-dependent impact strain levels at the resolution of single cells to determine the first, cell-based injury thresholds for traumatic brain injuries.

We fabricated neural cell networks following previously established 3D, cellular in vitro TBI model. Primary Sprague-Dawley rat cortices are isolated from p0-p1 rat pups. Isolated cortical cells are suspended in rat tail collagen-I hydrogel, cast into the 3D molds, and allowed to undergo synaptogenesis over seven days in vitro (DIV) prior to mechanical deformations. All samples are stained with Calcein AM and a 3D reference image at 7 DIV is acquired using a Nikon-A1 multiphoton microscope before samples are subjected to precisely controlled mechanical loadings, which are applied by one of three custom devices. Each device is capable of inducing one of the fundamental deformation types (compression, tension, shear) at rates up to 240s^-1 using state-of-the-art voice coil linear actuators. A high-speed image is acquired during the impact event using a high-speed camera and epifluorescent microscopy. A second volumetric image is taken at the same location immediately after the impact to assess immediate damage due to the insult (primary injury). The impact sample is then returned to the incubator and fixed at 24 hours post injury in 4% paraformaldehyde and 8% sucrose and immunolabeled for cytoskeletal and enzymatic markers of interest (e.g., actin and cleaved caspase-3). The cell population viability is assessed by tracking the cell morphology changes in the pre- and 24 hours post-injury images. Different cell types and distinct branch orientations result in heterogeneous strain and rate distributions. To fully resolve the cellular injury thresholds, we perform 3D neural cell segmentation followed by mapping the local impact strains onto the branched, 3D cell structures using an in-house developed MATLAB code package, allowing us to generate a statistical model that predicts injury at the resolution of a single cell. All existing cell types (i.e. neurons, astrocytes, etc.) are classified using a deep learning algorithm to resolve cell-type-specific injury susceptibility as well as an overall cellular injury risk curve. Through these detailed analyses, we are able to evaluate the strain magnitude, rate, and modality-dependent neuronal injury and resolve cellular injury thresholds and pathways. This approach has a broad application for in vitro and in situ mechanical impact experimentation and provides the first comprehensive analysis toolbox that resolves cellular TBI thresholds and could lead to the development of future generations of protective equipment.

 

Presenting Author: Christian Franck University of Wisconsin-Madison

Presenting Author Biography: Christian Franck is a mechanical engineer specializing in cellular biomechanics and new experimental mechanics techniques at the micro and nanoscale. He received his B.S. in aerospace engineering from the University of Virginia in 2003, and his M.S. and Ph.D. from the California Institute of Technology in 2004 and 2008. Dr. Franck held a post-doctoral position at Harvard investigating brain and neural trauma. He was an assistant and associate professor in mechanics at Brown University from 2009 - 2018, and is now the Grainger Institute for Engineering Professor in Mechanical Engineering at the University of Wisconsin-Madison.<br/>His lab at the University of Wisconsin-Madison has developed unique three-dimensional full-field imaging capabilities based on multiphoton microscopy and digital volume correlation. Current application areas of these three-dimensional microscopy techniques include understanding the 3D deformation behavior of neurons in the brain during traumatic brain injuries, and the role of non-linear material deformations in soft matter.<br/>He is the acting director of the Center for Traumatic Brain Injury at the University of Wisconsin-Madison and the ONR-funded Physics-based Neutralization of Threats to Human Tissues and Organs (PANTHER) program, which consists of over 24 PIs nationwide. Key objectives of the Panther program are in better detection, prediction, and prevention of traumatic brain injuries by providing accelerated translation from basic science discovery to civilian and warfighter protection solutions.

Authors:

Luke Summey University of Wisconsin-Madison
Annalise Daul University of Wisconsin-Madison
Jessica Park University of Wisconsin-Madison
Jamie Sergay University of Wisconsin-Madison
Jing Zhang University of Wisconsin-Madison
Christian Franck University of Wisconsin-Madison