Quantitative systems pharmacology model of axonal degeneration markers in Alzheimer’s disease
Tatiana Karelina, Polina Pchelintseva
InSysBio
Objectives: Progress in the treatment of neurodegenerative disease is limited by long duration of clinical trials required to establish clinical benefit of therapies. Is is supposed that analysis of biomarkers specifically related to neuronal and axonal damage would facilitate evaluation of therapy efficacy. One of the promising biomarkers is one of the neuronal cytoskeleton proteins, neurofilaments light chain (NfL). Its concentration is increased in plasma and cerebrospinal fluid (CSF) of patients with neurodegenerative diseases, including Alzheimer’s Disease (AD) and traumatic brain injury. Neurofilament lightis potentially useful in both AD diagnosis and for studying the preclinical stages of pathogenic progression [1]. They may also play specific role in pathology, as neurofilaments are contained in the neurofibrillary tangles, one of the hallmarks of AD [2]; neurons expressing neurofilaments are much more vulnerable to tau pathology and degeneration, neurofilament compactions (pathological aggregates of neurofilaments) are the earliest markers of axonal injury. Concentration of NfL correlates with rate of hippocampal atrophy and white matter hyperintensities (WMH), marker of axonal damage.
The goal is to develop translational quantitative systems pharmacology (QSP) model describing relationship between violation of intraneuronal processes and the change of the concentration of neurofilament L in the brain and cerebrospinal fluid during the progression of Alzheimer’s disease.
Methods: The model describes synthesis of neurofilaments and their degradation by calpain, proteasome and autophagic system; phosphorylation/dephosphorylation, formation of pathological aggregates, and the release of neurofilament proteins from degenerating axons into the extracellular space with distribution to the CSF. Description of the neurofilament degradation processes is partially based on neuron homeostasis model developed earlier [3]. To calibrate the model, we used literature data on the concentration of neurofilaments in brain and cerebrospinal fluid, the level of phosphorylation, and in vitro data on degradation of neurofilaments in different conditions (inhibition of calpain, proteasome or autophagy). Brain atrophy data are used to describe reduction of intracellular volume leading to additional excretion of neurofilaments into extracellular space. Tau and amyloid concentrations are introduced as explicit functions depending on age.
Results: Introduction of amyloid and tau toxic influence on neuronal transport and neurofilament degradation mechanisms allows for satisfactory description of the increase of neurofilament in CSF of AD subjects vs healthy subjects (about 1.5 times). Model describes accumulation of aggregated neurofilaments in the brain cells, corresponding to increase of WMH (about 2-3 times) [4]. However, to be able to describe much higher increase of NfL in CSF of amyloid transgenic APP/PS1 mouse [5], additional increase of NfL leakage to extracellular space was introduced. We assume that it may correspond to high level of membrane damage in this mouse and processes related to dystrophy of neurites. Model describes the reduction of NfL concentration in CSF upon inhibition of amyloid production in APP/PS1 mouse.
Conclusions: Previously amyloid and tau pathology progression and interaction was described through the interruption of intracellular processes [1]. Model presented here will help to connect previous results to other clinically important biomarkers, related to neurodegeneration. As axonal damage markers are common for different neurodegenerative diseases, the model can be integrated potentially with models of other pathologies (Parkinson’s disease, multiple sclerosis).
References:
[1] Alzheimers Dement (Amst), 2020, 12(1):e12005.
[2] Proc Natl Acad Sci U S A,1985, 82(11):3916-20.
[3] CPT: Pharmacometrics & Systems Pharmacology, December 2020, 1–8.
[4] Neurobiology of Aging, 2018, 68, 18–25.
[5] Neuron, 2016, 91(1), 56–66.