Study On Stochastic Dynamics In Intracellular Gene Regulatory Network With MicroRNA

The study of biomolecular interaction and its laws is an important field in modern molecular biology.Based on molecular biological experiments,systematic biology which studies the mechanism of intermolecular interaction is helpful to understand the nature of many important life phenomena such as complex diseases,life span of animals and plants,rhythm of life cycle and so on.The reconstruction and function prediction of gene regulatory networks in systems biology is one of the challenging topics.As an important class of non-coding RNAs in eukaryotes,microRNA(miRNAs)have been widely studied in many biological experiments.A large number of biological experiments have shown that microRNA inhibits the expression of target genes and play an important role in biological development,cell senescence and apoptosis,tomorigenesis and metastasis,so it is very important to study the dynamics of gene network with miRNA regulation in both experimentally and theoretically.Based on the existing research about the cell cycle and the gene regulatory network in the process of tumor cell EMT/MET,this paper mainly considers the effects of miRNA,biochemical noise and interaction time delay on these two biological processes,and then the corresponding nonlinear dynamics are established.The main results of this paper are as follows:(1)Considering the effect of miR-17-92 on the Rb-E2F gene regulatory network in cell cycle,results in formation of two positive feedback loops by Rb-E2F-CycE loop and E2F self activated loop,a E2F-miR-17-92 negative feedback loop,an incoherent feedforward loop by transcription factor Myc directly up-regulating the production of miR-17-92,ring the Rb-E2F-miRNA network by these four function loops.It's shown that the addition of miR-17-92 will greatly delay the G1 to S phase progression,cells can stay in a quiescent period longer.The delay effect induced by changing the inhibition rate of miR-17-92 on E2F is more obvious than changing the activation rate of miR-17-92 by Myc.The case of presence of noise,it's found that the addition of miR-17-92 can make cells to having a higher probability of being in the G1 phase,so as to maintain the robustness of the network.Based on the dynamic analysis of different regulatory pathway,we found that the addition of miRNA does not change the bistability.But changing the inhibition rate of miR-17-92 on E2F and activation rate of miR-17-92 by Myc will consequently lead to the miRNA "target avoidance" phenomenon,this phenomenon can be explained by the correlation analysis on expression of miRNA and E2F.It can as well provide a new theoretical explanation for miR-17-92 as a dual role player.(2)Based on the EMT regulatory network[miR-34/SNAIL]:[miR-200/ZEB],a reduced model was proposed.It's evinced that EMT and MET processes which were adduced in experiments can be reproduced.The asymmetry of E/M hybrid phenotype in the transformation of tumor cells is explained from the kinetic mechanism,that is,the E/M hybrid state only appears in the EMT process.By numerical simulation with noise,we found that under the tristable phenotypes,noise can lead to the transition of tumor cells from the E phenotype to M phenotype,and the hybrid phenotype have a certain ability to resist noise.In the presence of noise and time delay,it's found that noise can lead to high expression of ZEB and time delay weakens the effect of noise to keep ZEB at low expression level.There is a competitive effect in the presence of both noise and time delay.These conclusions may provide new ideas for clinical therapy of tumors.