Incorporating PCA to determine NMDAR Function in Neuronal Network Maturation – UROP Spring Symposium 2022

Incorporating PCA to determine NMDAR Function in Neuronal Network Maturation

photo of presenter

Vincent Wong

Pronouns: He/Him

Research Mentor(s): Kevin Jones
Co-Presenter:
Research Mentor School/College/Department: University of Michigan / Medicine
Presentation Date: April 20
Presentation Type: Oral10RS
Session: Session 5 – 3:40pm – 4:30 pm
Room: Vandenberg
Authors: Vincent Wong, Jean Rodriguez Diaz, Kevin Jones
Presenter: 4

Abstract

Schizophrenia is a psychiatric disease characterized by dysfunctional information processing ion the brain. Common symptoms include hallucinations, delusions, and abnormal behavior and speech patterns. The causes of schizophrenia are not known, but a subtype of ionotropic glutamate receptor known as N-Methyl D-Aspartate Receptors (NMDAR) are strongly implicated in the development of schizophrenia. Disrupting NMDAR function during early development causes lasting cognitive deficits in animal models that resemble schizophrenia. Whereas cognitive deficits of NMDAR block are well described, the impact of NMDAR block on neuronal network development is poorly understood. To determine how NMDAR dysfunction impacts the development of neuronal networks, we cultured rat hippocampal cultures on microelectrode arrays to record the electrical activity of neuronal networks that were transiently exposed to the NMDAR antagonist, MK-801. A preliminary analysis suggested MK-801 alters firing patterns within neuronal networks. However, the initial analysis only examined network-wide characteristics, which may have limited sensitivity. To improve the resolution and accuracy of the study we re-analyzed the data set using principal component analysis (PCA) and spike sorting to isolate single neuronal units. Re-analysis of the data suggests that patterned spiking emerges from vehicle-treated or MK-801-treated cultures at a similar rate. However, MK-801-treated cultures show a trend of developmentally delayed network synchrony. Our curent data set shows high variation In spike frequency between data sets which limits our interpretation. Nevertheless, this project demonstrated the efficacy and efficiency of incorporating PCA into our pipeline as a way of tracking spiking in individual neuronal units and network-wide bursting patterns. By demonstrating this proof-of-concept, this project advanced our ability to study schizophrenia in an NMDAR knockout model.

Presentation link

Biomedical Sciences, Natural/Life Sciences

lsa logoum logo