Techniques for Quantifying Structure-Function Relationships of Interstitial Cell of Cajal Networks

Abstract

Interstitial cells of Cajal (ICC) are specialised cells present throughout much of the gastrointestinal (GI) tract. These cells perform various functions to facilitate normal GI motility, with one of the most prominent roles being the electrical pacemaking of the GI tract. ICC loss and injury is now a major research focus as it is recognised as a hallmark of several GI functional motility disorders, but the mechanism relating ICC structure to GI function and dysfunction remain poorly defined. Progress in elucidating this mechanism is limited with experimental techniques alone due to factors including: 1) the absence of methods for quantifying the complex network structures formed by ICC; 2) challenges in associating cellular and tissue level activity across multiple spatial and temporal scales; and 3) the lack of a comprehensive ICC imaging data set encompassing large-scale network structures across a range of network properties. This thesis therefore aims to use a mathematical modelling approach to address these experimental shortcomings and investigate ICC network structure-function relationships. A set of six numerical metrics were developed to quantify the structural properties of confocal ICC network imaging data: density, thickness, hole size, contact ratio, connectivity and anisotropy. These metrics were applied to discern the effects of various gene knockouts (KO) and postnatal maturation (three-day- versus four-week-old) on murine intestinal ICC network structural properties, allowing for the first detailed automated analyses and unbiased quantitative comparisons of ICC network structures. The analysis revealed a novel remodelling phenomenon occurring during 5-HT2B KO (ICC depletion), namely a spatial rearrangement of ICC and the preferential longitudinal alignment of processes, and an apparent pruning of the ICC network occurring during postnatal maturation was identified as well. On the other hand, no changes in the ICC network structure were observed during Ano1 or Spry4 KO. The feasibility of employing multiscale computational models to relate ICC network structure to its electrical pacemaker activity was then evaluated as the models can be upscaled to span multiple spatial and temporal scales provided sufficient computational resources. First, to demonstrate utility, the established biophysically based simulation approach was used in conjunction with the developed ICC network structural metrics to investigate the structural and functional changes that occur in postnatal development (from birth to 24-day-old) of murine intestinal ICC networks. Four measures based on the average membrane potential and intracellular calcium concentration over the network ([Ca2+],) were used to quantitatively assess the simulated ICC pacemaker activity: activation rate, peak [Ca2+], time to peak [Ca2+]i and half peak [Ca2+], time ratio. The results identified a pruning-like mechanism occurring during postnatal ICC network development which may facilitate mature digestive function, and elucidated the temporal course of this developmental process. Next, a new cellular automaton model was constructed and used to demonstrate impaired pacemaker activity propagation during ICC depletion. This simulation approach represents a more simplistic but computationally-efficient option in comparison to biophysically-based simulations. Subsequently, an alternative strategy to experimental imaging for obtaining a comprehensive ICC network imaging data set encompassing large-scale ICC networks across a spectrum of network properties was presented. Both small- (0.225×0.225 mm with a resolution of 362×362 pixels) and large-scale realistic virtual ICC networks (≈25.2×12.4 mm with resolution of 40,500×20,000 pixels) were generated in silico using the stochastic Single Normal Equation Simulation (SNESIM) algorithm. The algorithm was also modified to enable virtual networks with a range of structural and functional properties to be generated. The SNESIM algorithm was then employed in conjunction with cellular automaton modelling to demonstrate the first multiscale computational framework in the GI field spanning from cellular structures to tissue level electrophysiology. The simulation results showed only a minor reduction in propagation velocities over 5-HT2B KO networks (4.3 mm/s) in comparison to WT networks (4.9 mm/s), which is logical as 5-HT2B KO does not affect intestinal transit times. In total, the work presented in this thesis constitutes novel mathematical and computational tools which form a comprehensive framework providing unprecedented analyses and a virtual platform for investigating ICC network structure-function relationships. Further refinement of this framework combined with concurrent improvements in computational resources will enable the elucidation of these key structure-function relationships, leading to a more extensive understanding of GI physiology and pathology, as well as holding promise for the clinical utility of ICC network information from patients to transform the diagnosis, prognosis and treatment of GI functional motility disorders.