Dr Rebecca Shipley, University College London, UK

Computational Modelling as a Tool to Direct the Design of Tissue-Engineered Peripheral Nerve Repair Constructs
When Feb 19, 2018
from 02:00 PM to 03:00 PM
Where LR8
Contact Name
Contact Phone 01865-283446
Add event to calendar vCal

The growth of nerves and blood vessels is intimately linked during development and repair, with both responding to a range of chemical and physical environmental cues. Understanding and controlling these cues is essential in many tissue repair scenarios, such as wound healing, grafting and the implantation of tissue-engineered constructs. Here we focus on peripheral nerve repair, where current clinical best practice is to bridge with a nerve autograft, inducing donor site morbidity and limited functional recovery. Although tissue engineering offers promising alternatives, these currently fall short of the autograft. This is due in part to a lack of understanding of how to organise therapeutic cells and materials within constructs to promote neuronal regeneration. The issue is compounded by the time, financial and ethical constraints of resolving these factors using purely in vitro and in vivo models.


Here we present an integrated framework of in vitro and in silico models to characterise the response of therapeutic cells to varying ambient oxygen conditions, representative of the physiological oxygen levels that would be experienced in a real repair scenario. A series of in vitro experiments explore how adipose-derived stem cells (embedded in compressed collagen gels at a range of seeded densities) alter their behaviour (viability, proliferation, growth factor production) when cultured at a range of oxygen conditions. These data are used to extract key parameters in a coupled partial differential equation model that describes the time and space evolution of cell, oxygen and growth factor concentrations within the same in vitro setup. Parameters are determined by formally minimising the disparity between experimental data and corresponding model predictions. Finally, the mathematical model is adapted to a simple representation of an in vivo repair geometry, with parameters from the in vitro experiments. Model predictions explore the link between spatial distribution of seeded cells, long-term cell survival and growth factor distributions. Virtual simulations are used to inform candidate cell seeding profiles that (a) minimise loss of valuable therapeutic cells while (b) inducing the growth factor cues required for vascularisation of the repair construct. This is a departure from traditional experimental approaches where cells are seeded uniformly at standard densities, and demonstrates the potential of a combined experimental-computational framework for improving the efficacy and speed of repair construct design.