Zhang Wei, Tian Tianhai, Zou Xiufen
School of Mathematics and Statistics, Wuhan University, Wuhan 430072, China; School of Sciences, East China Jiaotong University, Nanchang 330013, China.
School of Mathematical Science, Monash University, Melbourne Vic 3800, Australia.
Math Biosci. 2015 Jul;265:12-27. doi: 10.1016/j.mbs.2015.04.003. Epub 2015 Apr 17.
Type I interferon (IFN) signaling pathways play an essential role in the defense against early viral infections; however, the diverse and intricate molecular mechanisms of virus-triggered type I IFN responses are still poorly understood. In this study, we analyzed and compared two classes of models i.e., deterministic ordinary differential equations (ODEs) and stochastic models to elucidate the dynamics and stochasticity of type I IFN signaling pathways. Bifurcation analysis based on an ODE model reveals that the system exhibits a bistable switch and a one-way switch at high or low levels when the strengths of the negative and positive feedbacks are tuned. Furthermore, we compared the stochastic simulation results under the Master and Langevin equations. Both of the stochastic equations generate the bistable switch phenomenon, and the distance between two stable states are smaller than normal under the simulation of the Langevin equation. The quantitative computations also show that a moderate ratio between positive and negative feedback strengths is required to ensure a reliable switch between the different IFN concentrations that regulate the immune response. Moreover, we propose a multi-state stochastic model based on the above deterministic model to describe the multi-cellular system coupled with the diffusion of IFNs. The perturbation and inhibition analysis showed that the positive feedback, as well as noises, has little effect on the stochastic expression of IFNs, but the negative feedback of ISG56 on the activation of IRF7 has a great influence on IFN stochastic expression. Together, these results reveal that positive feedback stabilizes IFN gene expression, and negative feedback may be the main contribution to the stochastic expression of the IFN gene in the virus-triggered type I IFN response. These findings will provide new insight into the molecular mechanisms of virus-triggered type I IFN signaling pathways.
I型干扰素(IFN)信号通路在抵御早期病毒感染中起着至关重要的作用;然而,病毒触发的I型干扰素反应的多样且复杂的分子机制仍知之甚少。在本研究中,我们分析并比较了两类模型,即确定性常微分方程(ODE)模型和随机模型,以阐明I型干扰素信号通路的动力学和随机性。基于ODE模型的分岔分析表明,当正负反馈强度被调整时,系统在高水平或低水平时呈现双稳态开关和单向开关。此外,我们比较了主方程和朗之万方程下的随机模拟结果。这两个随机方程都产生了双稳态开关现象,并且在朗之万方程的模拟下,两个稳定状态之间的距离比正常情况小。定量计算还表明,需要正负反馈强度之间的适度比例来确保在调节免疫反应的不同干扰素浓度之间进行可靠的切换。此外,我们基于上述确定性模型提出了一个多状态随机模型,以描述与干扰素扩散耦合的多细胞系统。扰动和抑制分析表明,正反馈以及噪声对干扰素的随机表达影响很小,但ISG56对IRF7激活的负反馈对干扰素随机表达有很大影响。总之,这些结果表明正反馈稳定了干扰素基因表达,而负反馈可能是病毒触发的I型干扰素反应中干扰素基因随机表达的主要贡献因素。这些发现将为病毒触发的I型干扰素信号通路的分子机制提供新的见解。