A multiscale systems biology framework integrating ODE-based kinetics and MD-derived structural affinities to model mBDNF–proBDNF-mediated bifurcation dynamics in CNS neurotrophin signaling

Title : A multiscale systems biology framework integrating ODE-based kinetics and MD-derived structural affinities to model mBDNF–proBDNF-mediated bifurcation dynamics in CNS neurotrophin signaling

Abstract:

Brain-Derived Neurotrophic Factor (BDNF), a key member of the neurotrophin family, is implicated in the regulation of synaptic plasticity, neuronal differentiation, and long-term survival in the Central Nervous System (CNS). The dualistic signaling pathways mediated by the mature form of BDNF (mBDNF) via TrkB receptors and the precursor form (proBDNF) via p75NTR introduce a dynamic molecular axis that governs a balance between neurogenesis and programmed cell death.
In this study, we developed a multi-computational framework incorporating a system of a multitude of ordinary differential equations (ODEs) representing concentrations of proBDNF, BDNF, TrkB, p75NTR, tPA (cleavage enzyme), and the protein-receptor complexes. To parameterize the ODE system with biophysically accurate constants, Molecular Dynamics (MD) simulations were used to resolve energetically favorable conformations and receptor-binding affinities. Notably, these simulations enabled de novo structural elucidation of a previously uncharacterized isoform of proBDNF, increasing the biological relevance of the ODE model with heightened accuracy. A hill-type sigmoid function was then applied to the ODE system to create a nonlinear representation of the interaction between TrkB and p75 activation, translating directly to the topological rewiring of the neuronal model via modulation of edge weights for determination of overall neuronal death or survival.
This model provides insights towards emergent phenomena such as neurotrophin-dependent synaptic pruning, and it facilitates the identification of bifurcation thresholds in the proBDNF:mBDNF ratios that can signify early-stage neurodegeneration and can serve as biomarkers for various neurodegenerative diseases. The model also holds translational potential for therapeutic target discovery as well, and can be utilized to analyze the implications of different drugs on the neurotrophic ratio, hence providing further information on neurodegenerative states.

Biography:

Arpit Ramani is a 12th-grade student at Wheeler Magnet High School. For the past four years, he has conducted independent research on Alzheimer’s disease drug development, creating computational pipelines to model and evaluate therapeutic candidates. He also conducts research at Emory University using statistical genetics to identify biomarkers for Alzheimer’s disease. Arpit’s most recent work was done through the BU RISE program, in which he expanded his research by applying neuronal modeling to study brain health. Arpit hopes that his current efforts in Alzheimer’s disease research will serve as a foundation for his future pursuit of medicine and neurodegenerative drug development.