Contributions of cell behaviours to dorsal closure in Drosophila abdomen

尽管最近取得了进展，但对组织发育过程中细胞行为的调节和协调仍然知之甚少。我试图通过使用基于试剂的计算模型来解决这个问题，以调查不同的细胞行为在果蝇成体腹部表皮形态发生过程中对背部闭合的贡献程度。组织和器官的形成是形态发生过程的结果，形态发生过程依赖于多种细胞行为的调节和协调。在这个项目中，我将专注于果蝇(果蝇)成虫腹部表皮的发育，在这个过程中，幼虫上皮细胞(LECs)被成体组织母细胞取代[1]。活体4D显微镜(对活的果蝇的研究)为我们提供了收集数据的手段。多尺度数学模型使我们能够证明我们对组织内细胞动力学的理解，并使我们能够做出可以与数据进行比较的预测。然而，由于它们所涉及的假设，目前可用的模型只能提供有限的见解。这方面的一个例子是，大多数模型将单元表示为具有直边的多边形。然而，在形态发生过程中，已经表明细胞可以形成弯曲的、片状的突起，它们用来推动自己前进。为了研究这些突起对细胞迁移的影响程度，细胞迁移已被证明对成人表皮的正常闭合是必不可少的[1]，我的目标是扩展现有的模型，包括通过片状脂膜的迁移。我还打算在我的模型中加入不同的细胞死亡调控方式之间的区别。先前的一项研究发现，抑制成组织细胞的增殖导致LEC死亡的延迟，这表明存在一种协调这些过程的机制[2]。然而，这项研究并没有得出这种机制依赖于什么的结论。数学建模使我们有机会研究调节细胞死亡的不同事件之间的关系，即接触、机械信号和化学信号，以及[2]中确定的机制。在机械信号方面，数据表明，局部组织力学影响细胞分层的可能性(细胞从上皮中移除的一种方式)，这通常会导致细胞死亡[3]。先前的研究表明，大多数LEC直到成组织细胞巢(最初，成组织细胞聚集在“巢”中)开始扩张后才会死亡[1]。鉴于这一背景，我将调查这样一种假设，即由组织母细胞巢扩张引起的机械力会导致LEC分层，然后死亡。我将使用基于细胞的计算框架CHASTE[4]，因为这将使我能够对单个细胞的行为进行建模。从一个简单的模型开始，我将设计图像分析和推理算法来量化实验观察到的行为。然后，可以将其与模型结果进行比较，突出应该在哪里进行调整。我将继续这个模型修改和实验验证的循环，直到我的模型代表组织中细胞动力学的全部范围。果蝇的基因工具可以用来实验性地操纵这个系统，这使得不同的假设可以得到检验。通过操作单个模型参数，我可以比较这些结果，这将突出说明假设应该如何更新。模型和实验数据之间的这种相互作用将是至关重要的。我的模型将描述的复杂的组织重塑是高等生物体中类似现象的代表。因此，由于生物之间的高度遗传相似性，我开发的工具将与人类健康研究相关--特别是伤口愈合和肿瘤生长[6]，因为这些过程也依赖于细胞死亡和迁移行为的变化。

Despite recent progress, the regulation and coordination of cell behaviours during tissue development remain poorly understood. I seek to address this issue by using agent-based computational models to investigate the extent to which different cell behaviours contribute to dorsal closure during morphogenesis of the Drosophila adult abdominal epidermis. Shaping of tissues and organs arises as a result of morphogenetic processes, which rely on the regulation and coordination of a multitude of cell behaviours. In this project, I will be focusing on the development of the Drosophila (fruit fly) adult abdominal epidermis, during which larval epithelial cells (LECs) are replaced by adult histoblast cells [1].In vivo 4D microscopy (the study of live Drosophila) provides us with the means to collect data. Multi-scale mathematical models allow us to demonstrate our understanding of cell dynamics within a tissue and enable us to make predictions that can be compared to the data. However, due to assumptions that they involve, currently available models only provide limited insights. One example of this is that most models represent cells as polygons, which have straight edges. However, during morphogenesis, it has been shown that cells can form curved, lamellipodia-like protrusions that they use to propel themselves forwards. To investigate the extent to which these protrusions affect cell migration, which has been shown to be essential for the normal closure of the adult epidermis [1], I aim to extend existing models, by including migration via lamellipodia. I also aim to incorporate a distinction between the different ways in which cell death is regulated into my models. A previous study found that inhibiting histoblast proliferation led to a delay in LEC death, suggesting that there exists a mechanism which coordinates these processes [2]. However, the study did not conclude what this mechanism relies on. Mathematical modelling gives us the opportunity to investigate the relationship between the different events which regulate cell death, namely contact, mechanical signals and chemical signals, and the mechanism identified in [2].In terms of mechanical signals, data has shown that local tissue mechanics impact the likelihood that a cell will delaminate (a way that cells are removed from the epithelia), which usually then leads to cell death [3]. Previous research has shown that most LECs do not die until after the histoblast nests (initially, histoblasts are clustered together in 'nests') start to expand [1]. Given this background, I am to investigate the hypothesis that mechanical forces arising from histoblast nest expansion causes delamination and then death of LECs.I will be using the cell-based computational framework Chaste [4] as this will enable me to model individual cell behaviour. Starting with a simple model, I will design image analysis and inference algorithms to quantify behaviours observed experimentally. This can then be compared to the model outcomes, highlighting where adjustments should be made. I will continue with this cycle of model alterations and experimental validation until my model represents the full range of cell dynamics in a tissue. Genetic tools in Drosophila can be used to experimentally manipulate the system, which allows varying hypotheses to be tested. By manipulating individual model parameters, I can compare these results, which will highlight how the hypothesis should be updated. This interplay between model and experimental data will be vital.The complex reshaping of tissue that my models will describe is representative of similar phenomena in higher organisms. Therefore, due to the high genetic similarity between organisms, the tools I develop will be relevant to the study of human health - particularly wound healing [5] and tumour growth [6], as these processes also rely on both cell death and changes in migratory behaviour.