Synergistic control of oscillations in the NF-κB signalling pathway
Synergistic control of oscillations in the NF-κB signalling pathway
复制标题
DOI：
10.1049/ip-syb:20050050
复制
发表时间：
2005-09-01
期刊：
影响因子：
--
通讯作者：
Kell, DB
中科院分区：
文献类型：
--
作者：
Ihekwaba, AEC;Broomhead, DS;Kell, DB
在以前的工作中，我们研究了NF-κ B信号通路的一部分模型的行为。当模型参数变化时，模型显示出在数量、幅度和频率上都变化的振荡。灵敏度分析表明，只有9个的64个反应参数是主要负责控制的振荡时，这些参数是单独变化。然而，任何复杂系统的性质的控制是分布式的，并且，由于这些反应中的许多是高度非线性的，我们预计它们的相互作用也将是如此。这九个参数的成对调制给出了比所有64个反应的成对调制所需的搜索空间小约50倍（81对4096）的搜索空间，这使得他们的研究得以进行（否则这将是非常棘手的）。观察到显著的协同效应，其中一个参数的效果强烈（甚至定性）依赖于另一个参数的值。可以发现参数空间的区域，其中振荡的振幅而不是频率（定时）发生变化，反之亦然。这样的建模将允许设计和实验的性能，旨在解开振荡的动力学的作用，而不仅仅是他们的振幅，在确定细胞的命运。总的来说，分析揭示了这些动态模型的复杂程度，这是不明显的，从他们的个人参数单独的研究，并指出操纵复杂网络的多个元素，以实现所需的生理效果的价值。
In previous work, we studied the behaviour of a model of part of the NF-kappa B signalling pathway. The model displayed oscillations that varied both in number, amplitude and frequency when its parameters were varied. Sensitivity analysis showed that just nine of the 64 reaction parameters were mainly responsible for the control of the oscillations when these parameters were varied individually. However, the control of the properties of any complex system is distributed, and, as many of these reactions are highly non-linear, we expect that their interactions will be too. Pairwise modulation of these nine parameters gives a search space some 50 times smaller (81 against 4096) than that required for the pairwise modulation of all 64 reactions, and this permitted their study (which would otherwise have been effectively intractable). Strikingly synergistic effects were observed, in which the effect of one of the parameters was strongly (and even qualitatively) dependent on the values of another parameter. Regions of parameter space could be found in which the amplitude, but not the frequency (timing), of oscillations varied, and vice versa. Such modelling will permit the design and performance of experiments aimed at disentangling the role of the dynamics of oscillations, rather than simply their amplitude, in determining cell fate. Overall, the analyses reveal a level of complexity in these dynamic models that is not apparent from study of their individual parameters alone and point to the value of manipulating multiple elements of complex networks to achieve desired physiological effects.