Scientists studying amyotrophic lateral sclerosis (ALS), Parkinson's, Alzheimer's and other neurological diseases are getting unprecedented views inside brain cells, using powerful new microscopes to uncover changes that may begin long before neurons die.
These technologies enable researchers to zoom in on individual neurons and the tiny structures inside them, watch cells at work in real time and use artificial intelligence (AI) to identify patterns linked to disease, according to experts speaking at the Presidential Symposium – "Advances in Microscopy for Visualizing Neurologic Disease" – at the 151st Annual Meeting of the American Neurological Association (ANA).
"Neurological diseases ultimately arise from changes in cells and the extraordinarily complex structures within them. Until recently, many of those changes could not be visualized," said Dimitri Krainc, MD, PhD, FANA, president of the ANA, chair of the symposium, chair of the Davee Department of Neurology, director of the Feinberg Neuroscience Institute and Aaron Montgomery Ward Professor at Northwestern University Feinberg School of Medicine, Chicago. "These technologies are giving us a much more detailed and dynamic view of disease biology, with the potential to reveal abnormalities earlier and point us toward treatments that address the underlying molecular disruptions in neurons."
Super-Resolution Microscopy Reveals a New View Inside Neurons
Craig Blackstone, MD, PhD, FANA, is examining the endoplasmic reticulum (ER), an organelle network inside cells that performs several essential jobs. Advanced microscopy approaches reveal that parts once thought to be flat sheets are constantly changing networks of tiny tubes, providing a new view of how cells work and what may go wrong in disease, such as mutations in proteins that are linked to inherited neurological disorders, including hereditary spastic paraplegias.
"The ability to zoom in beyond the limits of conventional microscopy is changing the questions we can ask about neurological disease," said Dr. Blackstone, who is presenting "NeuroZoom: Super-Resolution for Neurologic Disease Discovery" and is the Julieanne Dorn Professor of Neurology at Harvard Medical School and inaugural scientific director of the Mass General Brigham Neuroscience Institute, Boston. "When we can see how structures inside neurons are organized and how they change dynamically in disease, we can begin to connect a genetic or molecular abnormality to what is happening inside the cell. That can reveal disease mechanisms we could not see before and, ultimately, new places to intervene."
That level of detail is especially relevant to neurological disorders because neurons are unusually large, highly specialized cells that must maintain their internal architecture over long distances and for a lifetime.
Advanced Imaging May Reveal Problems Inside Neurons Before Cells Die
Live imaging lets researchers watch organelles move and function in real time. They can observe how mitochondria produce and deliver energy and how cells use autophagy to remove damaged mitochondria and other large waste products, much like hauling away an old sofa that is too large for a regular trash bin.
"It's like magic, like watching a time-lapse video of a city at night," said Erika Holzbaur, PhD, who is presenting "Organelle Quality Control in Neuronal Homeostasis, Neurodevelopment, and Neurodegeneration" and is the William Maul Measey Professor of Physiology at the University of Pennsylvania Perelman School of Medicine, Philadelphia. "We can see the cell function in real time, see organelles appear, move and disappear. We can watch synapses fire and see how neurons react to dysfunctional mitochondria. These live movies provide clues that allow us to better understand the underlying mechanisms and how they are altered when there is disease."
Dr. Holzbaur will discuss how her team used live imaging to uncover an unexpected connection between Fragile X syndrome and mitochondria, which provide cells with energy. Studies in mouse and human neurons found that without FMRP, the protein missing in people with Fragile X, mitochondrial malfunctioned, pointing to possible new directions for treatment.
By the time a neuron dies in a patient with ALS, Parkinson's or Huntington's disease, it is too late to save it. We want to identify the first clues that these cells are stressed and better understand why. If we can do that, we have some hope of saving neurons before they degenerate and die."
Erika Holzbaur, William Maul Measey Professor of Physiology, Perelman School of Medicine, University of Pennsylvania
AI-Powered Microscopy May Help Identify ALS Sooner
ALS is notoriously difficult to diagnose because there is no definitive test for the disease. Diagnosis depends on a neurologist recognizing a pattern of symptoms and ruling out other conditions, and patients may spend close to a year seeing doctors and specialists before ALS is confirmed. That means losing valuable time to begin the few available treatments or enroll in clinical trials, which may be most helpful earlier in the disease.
Robotic microscopy paired with AI could eventually provide a faster, more objective way to support diagnosis. These "thinking" microscopes can track thousands of individual motor neurons made in the laboratory from patients' skin or blood cells, recording how each cell looks, functions and changes over time. Using AI, it can then analyze the massive collection of images to detect disease-related patterns that would be difficult or impossible for researchers to identify manually, as well as design and conduct experiments, learn from the results and decide what to test next.
"What's notable about our new finding is that an AI model, trained only on how patient-derived motor neurons look and behave under the microscope, could reliably tell ALS neurons apart from healthy ones," said Steven Finkbeiner, MD, PhD, who is presenting "State-of-the-Art Technologies to Generate and Analyze Large Amounts of Imaging and Genetic Data, Including Robotic Microscopy and AI." "That raises the possibility of offering a faster, more objective way to diagnose a patient when the clinical picture is unclear, rather than waiting for symptoms to unfold over months."
AI also sorted the ALS neurons into meaningful subtypes and traced one subtype to a problem in how proteins move in and out of the cell's nucleus, potentially providing clues to what drives the disease. In separate research, Finkbeiner's team used AI to automatically identify and measure signs of Alzheimer's disease in postmortem brain tissue that tracked with dementia severity, meaning researchers can analyze large collections of brain samples more quickly.
"This approach isn't specific to one disease - the same imaging and AI tools can be used across ALS, Alzheimer's, Parkinson's, Huntington's and frontotemporal dementia," Finkbeiner said. "This work is in its early stages, but a technology developed to answer a question in one disease often turns out to accelerate discovery in several others."
Study Challenges Traditional Nutrition Assumptions Regarding Equivalent Dietary Calorie Impacts