Brain-on-a-chip microsystem for investigating traumatic brain injury: Axon diameter and mitochondrial membrane changes play a significant role in axonal response to strain injuries

Abstract

Diffuse axonal injury (DAI) is a devastating consequence of traumatic brain injury, resulting in significant axon and neuronal degeneration. Currently, therapeutic options are limited. Using our brain-on-a-chip device, we evaluated axonal responses to DAI. We observed that axonal diameter plays a significant role in response to strain injury, which correlated to delayed elasticity and inversely correlated to axonal beading and axonal degeneration. When changes in mitochondrial membrane potential (MMP) were monitored an applied strain injury threshold was noted, below which delayed hyperpolarization was observed and above which immediate depolarization occurred. When the NHE-1 inhibitor EIPA was administered before injury, inhibition in both hyperpolarization and depolarization occurred along with axonal degeneration. Therefore, axonal diameter plays a significant role in strain injury and our brain-on-a-chip technology can be used both to understand the biochemical consequences of DAI and screen for potential therapeutic agents.

Figure 1

Organotypic uniaxial axonal strain device. (a) Schematic of device before and after strain application. (b) Detailed top view of assembled device. (c) Strain profile (E_xx) across the pressure cavity width. a and b are not to scale.

Figure 2

Uniaxial strain injury induced axonal beading. (a) Example of axonal beading (arrows) observed after uniaxial strain injury, (i) before injury and (ii) 4 hours after 45% strain injury. (b) The number of beads occurring along the length of an axon within 4 hours after injury normalized per 100 µm length for 10%, 25% and 45% applied strain. *P < 0.05 compared to 10% strain. (c) Diameter dependence of axonal beading after 10%, 25% and 45% uniaxial strain injury. The diameters of axons were grouped into 0.4 µm bins from 0.6 µm to 2.6 µm and the number of beads occurring along the length of an axon is normalized per 100 µm length. *P < 0.05 compared to 0.6 µm–1.0 µm for each specific applied strain. All error bars represent s.e.m. (number of experiments = 6–10). Scale bar, 10 µm.

Figure 3

Increase in axon length following uniaxial strain injury. (a) Example of undulations following a 45% strain injury, (i) before injury and (ii) immediately after injury. (b) Increase in axon length due to strain for 10%, 25% and 45% applied strains. *P < 0.05 compared to 10% strain (number of experiments = 9). (c) The effect of different diameters on the increase in axon length following a 10%, 25% and 45% uniaxial strain injury. The diameters of axons were grouped into 0.4 µm bins from 0.6 µm to 3.0 µm. (d) The number of undulations observed along the axon length per 100 µm after 10%, 25% and 45% uniaxial strain injury. (e) The amplitude of undulations occuring along the length of the axon (normalized to the axon diameter before injury) after 10%, 25% and 45% uniaxial strain injury. (f) The time taken for axons to return to their original pre injury length after 10%, 25% and 45% uniaxial strian injury. *P < 0.05 compared to 0.6 µm–1.0 µm for c to f (number of experiments = 9). Scale bar, 10 µm.

Figure 4

Axonal degradation following a uniaxial strain injury. (a) Axonal degradation was assessed at 24 hours following uniaxial strain injury with respect to 10%, 25% and 45% applied strain. *P < 0.05 compared to 10% strain (number of experiments = 12). (b) The time taken to degrade was monitored with respect to axon diameter. The diameters of axons were grouped into 0.4 µm bins from 0.6 µm to 3.0 µm. *P < 0.05 compared to 0.6 µm-1.0 µm (number of experiments = 12). (c) An example of an axon along the full length of the strain region undergoing degradation over time. Three different positions are highlighted and shown in higher magnification in d–f. Position d and f correspond to the regions of lowest strain and e the region of highest strain. (d–f) (i) axon before injury, (ii) axon immediately following injury, (iii) 18 hours post injury and (iv) 20 hours post injury.

Figure 5

Monitoring MMP changes over a 24-hour period following uniaxial strain injury. MMP changes are normalized to their potential before injury and assessed at 11 discrete sections along the axons and 6 time points, i.e. immediately following injury (0 hour), 1 hour, 2 hours, 4 hours, 9 hours and 24 hours post injury. (a) Uninjured control and addition of mitochondrial uncoupler FCCP. Changes in MMP after applying a uniaxial strain injury, (b) 10% applied strain, (c) 25% applied strain and (d) 45% applied strain. MMP changes were assessed after application of the NHE-1 inhibitor EIPA and uniaxial strain injuries. (e) EIPA and 10% applied strain, (f) EIPA and 25% applied strain and (g) EIPA and 45% applied strain. *P < 0.05 compared to MMP at that particular position before injury. ^†P < 0.05 compared to MMP at that particular time point at positions 1 and 11 (number of experiments = 8).

Figure 6

Axonal degradation assessed at 24 hours following uniaxial strain injury after treatment with EIPA and CsA. Degradation was assessed for three uniaxial strains, i.e. 10%, 25% and 45%. *P < 0.05 compared to 10% strain, ^†P < 0.05 compared to control (untreated) for that particular applied strain injury (number of experiments = 4–12).