Development of a novel in silico model for the undiseased zebrafish electrophysiology

Heart Failure (HF) is a complex pathophysiological syndrome characterized by the impaired ability of the heart to intake and/or eject sufficient blood, leading to significant mortality, disability, and healthcare expenditure. Affecting approximately 2.8% of the adult population, it represents a predominant cause of mortality in Western countries. The limited availability of human cardiac tissue and the complexity of patient phenotypes make the study of these pathologies highly challenging. Consequently, the use of experimentally adaptable animal models has become crucial in contemporary research. In recent decades, the zebrafish (Danio rerio) has emerged as an exceptional genetic and embryonic animal model for investigating HF. Despite this interest, no efforts have been made to develop numerical models of the zebrafish to date. This could be related to the difficulties in finding available images of the entire heart and body in different developmental stages and adult fishes, but it could also be associated with the lack of specific data regarding the ionic currents involved in the zebrafish action potential and their ionic channels. Therefore, this research aims to fill this gap through the development of an electrophysiological in silico model of zebrafish, including both heart and body, capable of enhancing our comprehension of the ionic mechanisms implicated in the progression of cardiac pathologies, such as arrhythmias. Furthermore, it pursuits to investigate the effects of drugs administered in relation to these pathological conditions, elucidating their influence on the action potential morphology and, consequently, on the associated electrocardiogram (ECG). To accomplish this objective, computational work supported by experimental activities was carried out in this research, leading to the development of an electrophysiological Finite Element Model (FEM) and electrophysiologically detailed numerical models for atrial and ventricular Action Potentials (APs). In both the FEM and AP models, the response to specific channel blockers was incorporated to assess their capabilities in relation to drug administration. Finally, with the aim of making the model increasingly comprehensive, a preliminary study was conducted regarding the coupling of the new electrophysiological model with mechanics, thereby introducing cardiac contraction. Particularly, the first step in the development of the zebrafish digital twin, as discussed in Chapter 3, involves the development of a finite element model for a 3 days post-fertilization (dpf) zebrafish embryo including the whole body and the two chambers of the heart (i.e., atrium and ventricle). A four-variable phenomenological action potential model describes the action potential of the different heart regions. Tissue conductivity is calibrated to replicate the experimentally described activation sequence. Moreover, the inclusion of the body enables the extraction of both monopolar and bipolar ECG traces to be compared with experimental ones reported in the literature. Secondly, for gaining insights into ionic channels, the use of a detailed action potential model instead of a phenomenological one is crucial. For this reason, Chapter 4 focuses on the development of a detailed AP model for both the ventricle and the atrium. The ventricular model is derived by reparametrizing the existing human model from TenTuscher and Pafilov (2004) due to the lack of experimental data regarding plateau and background currents, as well as the pump and exchangers. This reparametrization is also guided by the similarities between zebrafish and human electrophysiology. The model is validated using experimental recordings conducted under the same stimulation protocols used for the numerical model (i.e., steady state, S1S2, and dynamic protocols). Moreover, the response of the developed model to specific channel blockers (i.e., Chromanol 239B, E-4031, and Quinidine) was assessed using a pore block model, and the results were compared with the literature. The preliminary atrial model, lacking experimental patch-clamp data available in the literature, is obtained through reparametrization of the newly developed ventricular model and is similarly validated through experimental AP and calcium transients performed under the same protocol used for the numerical model. In Chapter 5, the full potential of the in silico model is finally demonstrated by integrating the electrophysiological finite element model with the detailed action potential models. Similarly to what was done in Chapter 4, the response to specific channel blockers, such as Chromanol 239B, E-4031, and Quinidine, was evaluated using a pore block model. In this case, the assessment not only includes the analysis of the AP duration but also considers the prolongation of QT intervals in the ECG tracing, which is made possible by the incorporation of the body. Lastly, in Chapter 6, a preliminary exploration of coupling the electrophysiological model with mechanics is presented. This approach assesses the contraction of the model. This interdisciplinary approach, combining computational modeling supported by experimental validation, contributes to advancing our understanding of zebrafish heart electrophysiology, with potential implications for studying cardiac pathologies and improving drug testing methodologies. Additionally, the developed model seeks to significantly reduce the number of animals used in vivo experiments by refining experimental procedures.