Scientists at Johns Hopkins University have developed a groundbreaking technique utilizing artificial intelligence to visualize and monitor changes in the strength of synapses, which are vital connection points facilitating communication between nerve cells in the brain. Published in Nature Methods, this innovative approach is expected to deepen our understanding of how synapse connections evolve in response to learning, aging, injury, and disease.
The method, devised by Dwight Bergles, Adam Charles, Jeremias Sulam, and Richard Huganir from Johns Hopkins’ Kavli Neuroscience Discovery Institute, allows for the observation of individual synapses in the brains of live animals. By closely monitoring synapse dynamics over time, the researchers aim to uncover crucial insights into brain function and the underlying mechanisms behind synaptic changes.
Traditionally, visualizing the shifting chemistry of synaptic messaging has been a challenge due to the high density and small size of synapses. To address this, the team turned to machine learning, leveraging its ability to enhance images composed of thousands of synapses. Machine learning algorithms were trained to improve image quality, enabling the detection and tracking of individual synapses during multiday experiments.
The researchers worked with genetically modified mice whose glutamate receptors emitted fluorescence when activated by light, providing an indication of synaptic strength. By capturing repeated images of the same synapses in live mice over several weeks, the team observed changes in glutamate receptor levels following exposure to new environmental stimuli. The experiments revealed a spectrum of alterations in synapse fluorescence, indicating both strengthening and weakening connections in response to environmental changes.
The interdisciplinary collaboration between scientists specializing in molecular biology and artificial intelligence was key to the success of this research. The Kavli Neuroscience Discovery Institute promotes such partnerships, facilitating breakthroughs at the intersection of different fields.
Building upon their findings, the scientists are now applying their machine learning approach to study synaptic changes in animal models of Alzheimer’s disease. They anticipate that this method will shed new light on synaptic alterations occurring in various disease and injury contexts.
The impact of this research extends beyond the lab, as it holds the potential to revolutionize our understanding of brain function and pave the way for targeted interventions in conditions involving synaptic dysfunction.
