A 3D Culture Model to Investigate Cellular Responses to Mechanical Loading in Spinal Cord Injury

Spinal cord injury (SCI) can cause paralysis, loss of sensation, and respiratory dependency, which has a significant impact on the quality of life of patients, their life expectancy and is also a significant economic burden due to the high costs associated with primary care and loss of income. One of the difficulties in establishing a treatment method is the heterogeneity of SCI; there are many different types and severities of traumatic primary injury, across different age groups of patients and different locations within the spinal cord, whilst at a cellular level, there are multiple, interacting secondary injury cascades that amplify the primary damage inflicted during the traumatic insult. Many techniques have been developed to mimic particular injuries found in human SCI, however in vivo animal models can be extremely costly and time consuming. The modest translation of therapeutic treatments from animal models to successful clinical trials suggests that there is a need for simplified models of SCI, in which the complex secondary cascade can be broken down into specific cellular interactions under controlled injury parameters. It was hypothesised that in vivo injuries could be simulated using a 3D in vitro model of SCI within a tethered, self-aligned, type-I collagen gel. An in vitro model such as this could advance the understanding of cellular responses to injury and help inform animal studies which may facilitate the design of therapeutics. Initially, different matrices were investigated in order to determine their suitability for use as matrix components for a 3D in vitro model of SCI. The matrices were characterised in terms of their mechanical properties, and the cellular responses of astrocytes following culture within the matrices. A fully hydrated matrix was selected which had a lower elastic modulus in comparison to spinal cord tissue, and which maintained astrocytes in a non-reactive state, as determined by the expression of markers for reactive astrogliosis. Contusion models of SCI are thought to generate the most relevant animal models of SCI, therefore their suitability as an injury mechanism within a 3D cellular model was investigated. A pilot study using the Hatteras contusion device, demonstrated that there was potential for in vivo type contusion devices to be utilised with an in vitro 3D collagen gel SCI model. The remainder of the study utilised the Infinite Horizons (IH) in vivo impactor, which is a force controlled contusion device. The experimental parameters utilised with the IH impactor within an in vivo setting were investigated as to their suitability for collagen gel impactions. Following a detailed investigation, the in vivo parameters of an impact force of 200 kdyn and a dwell time of 0 ms, using a 2.5 mm diameter impaction tip were adopted; however the calibration start height of the impaction tip was altered to avoid full penetration of the impactor tip through the gel. The limitations of the contusion device affected the consistency of the impaction and resulted in a lack force output data. These limitations need to be resolved in order to directly compare in vivo and in vitro SCI using the IH impactor. The impaction of 3D aligned, collagen gels, seeded with primary rat astrocytes, using the IH impactor generated a 3D cavity bordered by reactive astrocytes, which was reminiscent of the glial scar and cystic cavity which forms at the lesion site in vivo. An increasing gradient of the astrocyte reactivity marker, glial fibrillary acidic protein, was expressed by cells closest to the impact zone. Astrocytes within the first 100 µm of the impact zone were highly ramified with cellular filaments aligned with the edge of the impact zone. An increase in the expression of astrocyte reactivity markers was observed over a ten-day period following impaction. In summary, a 3D model of SCI was developed that was highly adaptable, and suitable for further advancement to increase the complexity and experimental outputs that were presented in this study. More detailed analysis of the cellular responses, over longer time courses, and perhaps with the additional complexity of multiple cell types would complement investigations within in vivo models. 3D in vitro tethered collagen gel models such as this could provide valuable insights into the cellular mechanisms which may progress the translation of treatments into the clinic.