Fast-Slow Variable Dissection with Two Slow Variables: A Case Study on Bifurcations Underlying Bursting for Seizure and Spreading Depression

The identification of nonlinear dynamics of bursting patterns related to multiple time scales and pathology of brain tissues is still an open problem. In the present paper, representative cases of bursting related to seizure (SZ) and spreading depression (SD) simulated in a theoretical model are analyzed. When the fast-slow variable dissection method with only one slow variable (extracellular potassium concentration, [K+](0) taken as the bifurcation parameter of the fast subsystem is used, the mismatch between bifurcation points of the fast subsystem and the beginning and ending phases of burst appears. To overcome this problem, both slow variables [K+](o) and [Na+](i); (intracellular sodium concentration) are regarded as bifurcation parameters of the fast subsystem, which exhibits three codimension-2 bifurcation points and multiple codimension-1 bifurcation curves containing the saddle-node bifurcation on an invariant cycle (SNIC), the supercritical Hopf bifurcation (the border between spiking and the depolarization block), and the saddle homoclinic (HC) bifurcation. The bursting patterns for SD are related to the Hopf bifurcation and the depolarization block while for SZ to SNIC. Furthermore, at the intersection points between the bursting trajectory and the bifurcation curves in plane ([K+](o) , [Na+](i)), the initial or termination phases of burst match the SNIC or IIC point well or the Hopf point to a certain extent due to the slow passage effect, showing that the fast-slow variable dissection method with suitable process is still effective to analyze bursting activities. The results present the complex bifurcations underlying the bursting patterns and a proper performing process for the fast-slow variable dissection with two slow variables, which are helpful for modulation to bursting patterns related to brain disfunction.

Automated summary

The study explores bursting patterns related to seizures and spreading depression in a theoretical model. By utilizing a fast-slow variable dissection method with two slow variables, the analysis reveals complex bifurcations underlying the bursting patterns and a proper performing process for effective modulation to brain dysfunction-related bursting patterns.