Isolation of neuron-specific, biomarker-loaded extracellular vesicles paves way to next-generation precision diagnostics and fundamental discoveries on neurological disorders
By Benjamin Boettner
(BOSTON) — Devastating neurological and neuropsychiatric disorders like Parkinson’s disease, Alzheimer’s disease, ALS, and many others are only diagnosed when their first neurological or behavioral symptoms appear. At that time, their underlying pathological processes in the brain and other nervous tissues are already well advanced. Therefore, diagnostic biomarkers that coincide with early stages of neurological disorders can help track their progression and can determine responses to therapeutic drugs are highly desirable. However, their discovery has been greatly hindered by the brain’s limited accessibility to biopsy, prompting researchers to look elsewhere.
For over 10 years, researchers have investigated the possibility of taking a non-invasive “snapshot” of the state of neuronal cells (neurons) in patients’ brains and nervous systems by measuring the contents of nanoparticles called extracellular vesicles (EVs). EVs are membrane-enclosed particles containing RNA transcripts and proteins that are released by all cells into biological fluids such as blood, which makes them a rich source of potential biomarkers. However, isolating neuron-specific EVs has presented an extremely difficult challenge.
Now, building on a body of previous work, a research team at the Wyss Institute at Harvard University, Boston’s Brigham and Women’s Hospital, Broad Institute of MIT and Harvard, and Harvard Medical School (HMS), led by Wyss Core Faculty members David Walt, Ph.D., and George Church, Ph.D., overcame a vital bottleneck toward solving this challenge. By building a broadly applicable framework that integrates new EV isolation methods with omics and computational analysis, and cutting-edge ultra-sensitive detection methods, they identified proteins that can be targeted to isolate neuron-specific EVs using immunological antibody-based methods. Their proof-of-concept study is published in Nature Neuroscience.
The fact that neuron-specific EVs make up only a very small percentage of what is present in plasma and other biofluids, required us to find the literal needle in the haystack. However, the framework that emerged from this work has so much broader potential, for identifying EV markers for subtypes from other cells of the brain, and essentially any other tissue in the body.
“A few years ago, the field thought that with the marker L1CAM the problem of isolating neuron-specific EVs was solved, but we found that L1CAM was not on EVs, and therefore cannot be used. What came next was the veritable tour de force reported in this new study in which we hunted down the first viable markers with utility for isolating neuron-specific EVs,” said Walt. “The fact that neuron-specific EVs make up only a very small percentage of what is present in plasma and other biofluids required us to find the literal needle in the haystack. However, the framework that emerged from this work has so much broader potential for identifying EV markers for subtypes from other cells of the brain, and essentially any other tissue in the body.” Walt is also the Hansjörg Wyss Professor of Biologically Inspired Engineering at Harvard Medical School and Professor of Pathology at Brigham and Women’s Hospital.
“We reasoned that cell-type-specific EV surface protein markers had to fulfill three important criteria: they need to be present on EVs in the biofluid of interest, in our case blood plasma (plasma) or cerebrospinal fluid (CSF), be transmembrane proteins so that an antibody can pull at them at the EV surface while keeping EVs intact, and be specifically expressed in neuronal cells,” said Wyss Senior Scientist and co-corresponding first author Dmitry Ter-Ovanesyan, Ph.D., who spearheaded the study. “This framework is conceptually simple, but it turns out that isolating cell type-specific extracellular vesicles is highly challenging.”
Building an EV isolation pipeline
To be able to focus on cell type-specific EVs, the team first developed a new method to isolate the total spectrum of EVs from plasma and other body fluids to extremely high purity. Relative to EVs, the amounts of free proteins, like albumin in blood, are much higher, which inevitably contaminates EV preparations and makes it difficult to analyze EV contents. Using insights from their previous advances that already considerably reduced the amounts of free proteins in their preparations, the researchers established a new method in which porous beads are added to samples that trap and help eliminate proteins such as albumin but exclude larger particles like EVs. This increased the purity of EVs by another 100-fold in a simple one-step procedure.
Leveraging this method as well as the proteomics expertise of Bogdan Budnik, Ph.D., Director of Biomarker and Drug Target Discovery at the Wyss, the team used mass spectrometry to analyze the protein content from EVs isolated using the new high-purity method and created comprehensive reference proteomes for both plasma and CSF EVs – a compendium of proteins that can be found in EVs.
“From here, we needed to home in on both transmembrane proteins that traverse the outside of the EV membrane and, among those, proteins that are specifically expressed in neuronal cells. To help us overlay our reference EV proteome data with human gene expression data, we collaborated with computational biologists in Aviv Regev’s group,” said Ter-Ovanesyan. Co-corresponding author Regev, then working with her group at the Broad Institute and MIT, is a computational biologist and one of the pioneers of single cell analysis using novel computational approaches. She is now Head and Executive Vice President at Genentech’s Research and Early Development unit. The team used a new computational method developed in Regev’s group to analyze gene expression across many single cell RNA-Seq studies to help prioritize predicted transmembrane proteins that were expressed uniquely in neurons and not in other cells of the body.
“We developed an unbiased computational pipeline to rank cell type-specific EV markers and applied it to finding candidate neuronal EV markers. We then developed immunoassays to several of these markers and went back to confirming their presence on EVs from plasma,” said Ter-Ovanesyan. The researchers were able to confirm the presence of one candidate marker, NRXN3, on EVs from both CSF and plasma using both ultra-sensitive SIMOA detection assays and single EV imaging methods.
Into the EV future
Finally, to facilitate EV analysis in biofluid samples, the researchers developed and validated a straightforward, highly efficient and specific technique for immuno-isolating EVs. Applied to NRXN3, they were able to reliably immuno-isolate NRXN3-carrying EVs from both human CSF and plasma samples. “This project is a significant step towards our goal of reading out the brains of patients through the contents of neuron-specific EVs,” said Walt. “We are now measuring specific proteins inside of these EVs to confirm their neuronal origin. Our goal is to look inside neuron-derived EVs for markers that are correlated with earlier stages and the progression of neurological disorders, as well as therapeutic efficacy of present and future drug candidates. More fundamentally, such markers can also help us better understand some of the underlying early pathologies, which for many neurological disorders are not clear at all.”
Our goal is to look inside neuron-derived EVs for markers that are correlated with earlier stages and the progression of neurological disorders, as well as therapeutic efficacy of present and future drug candidates.
“The highly innovative and systematic way of how this Wyss team is developing extracellular vesicles as a diagnostic platform with potential for many diseases and disorders is a prime example of how our scientists attack daunting problems head-on, ‘peeling the skin of the onion,’ until, in the end, they provide solutions that not long ago where out of reach,” said Wyss Founding Director Donald Ingber, M.D., Ph.D. who is also the Judah Folkman Professor of Vascular Biology at Harvard Medical School and Boston Children’s Hospital, and the Hansjörg Wyss Professor of Biologically Inspired Engineering at Harvard John A. Paulson School of Engineering and Applied Sciences.
Other authors on the study were Sara Whiteman, Tal Gilboa, Emma Kowal, Wendy Trieu, Siddharth Iyer, Clarissa May Babila, Chungmin Han, David Vance, Graham Heimberg, Michael Burgess, Hasmik Keshishian, and Steven Carr. The study was funded by the NIH Center for
Excellence in Genomics Science, Howard Hughes Medical Institute, Klarman Cell Observatory, CZI Neurodegeneration Challenge Network, Open Philanthropy Project/Good Ventures, and Wellcome Leap as part of the Multi-Channel Psych Program.