A general mathematical framework to model generation structure in a population of asynchronously dividing cells
A general mathematical framework to model generation structure in a population of asynchronously dividing cells
复制标题
DOI：
10.1016/j.jtbi.2004.04.011
复制
发表时间：
2004-08-21
影响因子：
2
通讯作者：
Carneiro, J
中科院分区：
文献类型：
--
作者：
León, K;Faro, J;Carneiro, J
在同质细胞群中，单个细胞经历异步细胞周期。近年来，人们对这一基本观察的兴趣因 USE 的广泛使用而增加，USE 是一种荧光染料，可以通过流式细胞术精确估计群体中不同细胞的分裂次数，从而精确估计世代结构。在这项工作中，我们提出了两个通用数学框架来模拟细胞群中世代结构的时间演化。第一个模型框架更具描述性，并假设由于单个细胞中分子机制的固有噪声，细胞分裂时间分布在细胞群中；而第二个框架则假设细胞分裂的异步性源于个体细胞与环境因素相互作用的随机性。我们减少这些形式主义以恢复两个预先存在的模型，它们建立在每个假设的基础上。当面对从文献中获取的 CFSE 标记细胞的动力学数据时，这些模型可以拟合每个测量时间点的前体频率分布。然而，它们无法拟合前体频率分布的整个动力学。相比之下，这些模型的两个扩展（也源自我们的一般形式主义）同样适合每个时间点的整个动力学和个体曲线，从而提供了生物学上合理的参数估计。我们证明细胞分裂时间的分布不是如之前提出的高斯分布，而是通过非对称分布（例如伽玛分布）更好地描述。我们还表明，观察到的细胞异步可以通过细胞分裂过程中单个过渡事件的存在来解释。基于这些结果，我们提出了结合理论和实验工作的新方法，以评估细胞内部机制中的噪音以及与环境因素的相互作用有多少导致细胞分裂的异步。 (C) 2004 Elsevier Ltd. 保留所有权利。
In otherwise homogeneous cell populations, individual cells undergo asynchronous cell cycles. In recent years, interest in this fundamental observation has been boosted by the wide usage of USE, a fluorescent dye that allows the precise estimation by flow cytometry of the number of divisions performed by different cells in a population, and thus the generation structure. In this work, we propose two general mathematical frameworks to model the time evolution of generation structure in a cell population. The first modeling framework is more descriptive and assumes that cell division time is distributed in the cell population, due to intrinsic noise in the molecular machinery in individual cells; while the second framework assumes that asynchrony in cell division stems from randomness in the interactions individual cells make with environmental agents. We reduce these formalisms to recover two preexistent models, which build on each of the hypotheses. When confronted to kinetics data on CFSE labeled cells taken from literature, these models can fit precursor frequency distributions at each measured time point. However, they fail to fit the whole kinetics of precursor frequency distributions. In contrast, two extensions of those models, derived also from our general formalisms, fit equally well both the whole kinetics and individual profiles at each time point, providing a biologically reasonable estimation of parameters. We prove that the distribution of cell division times is not Gaussian, as previously proposed, but is better described by an asymmetric distribution such as the Gamma distribution. We show also that the observed cell asynchrony could be explained by the existence of a single transitional event during cell division. Based on these results, we suggest new ways of combining theoretical and experimental work to assess how much of noise in internal machinery of the cell and interactions with the environmental agents contribute to the asynchrony in cell division. (C) 2004 Elsevier Ltd. All rights reserved.