New Sequencing Methods Capture the Brain’s Hidden Vascular Cells

The blood vessels of the human brain have long been among the most consequential and least accessible components of neurology. Endothelial cells lining vessel walls, mural cells such as pericytes and smooth muscle cells, and an array of perivascular immune populations together form the cerebrovasculature, the living pipeline that sustains every neuron in the brain. When these cells falter, the consequences are severe: vascular dysfunction is now recognized as a central driver of stroke, vascular dementia, and Alzheimer’s disease. Yet despite their importance, these cells have remained stubbornly invisible to the single-cell genomics revolution that has transformed the study of neurons and glia over the past decade. A newly published protocol in Nature Protocols aims to close that gap by providing scientists everywhere with a detailed, reproducible recipe for capturing the brain’s vascular cells at single-cell resolution.

The challenge that the method, called VINE-seq, was designed to solve is fundamentally mechanical. Most single-nucleus sequencing workflows begin by gently breaking brain tissue apart and releasing nuclei from individual cells. That approach works reasonably well for neurons and glia, but vascular cells are a different story. They are encased within a tough basement membrane, a dense meshwork of proteins that resists standard dissociation methods. As a result, when researchers homogenize brain tissue for single-nucleus profiling, vascular cells are systematically lost, depleted, or damaged. The consequence has been a major blind spot in the single-cell atlases of the human brain that now underpin much of modern neuroscience: the very cells that form the blood-brain barrier and regulate cerebral blood flow are dramatically underrepresented in the datasets used to study health and disease.

VINE-seq, short for vessel isolation and nucleus extraction for sequencing, tackles the problem head-on by isolating the vessels themselves before any attempt is made to extract their genetic material. The workflow begins with fresh or frozen brain tissue, either human or mouse, which is homogenized under conditions that preserve vascular integrity. The homogenate is then subjected to dextran-based density-gradient centrifugation, a technique that exploits differences in buoyant density to separate intact vessels from the surrounding myelin and the parenchymal fraction containing neurons and glia. The vessels recovered by this step are predominantly capillaries and small arterioles and venules, roughly 100 micrometers in diameter or smaller, which represent the functional workhorses of the cerebral circulation.

Once the vascular pellet has been collected, the protocol calls for rigorous washing of the isolated vessels over a cell strainer. This deceptively simple step removes trapped contaminants, including parenchymal cells and debris that can cling to the outside of vessel fragments and muddy downstream molecular profiles. The critical innovation, however, comes in the third stage: the optimized extraction of nuclei from the purified vessels using enzymatic digestion, specifically with collagenase III. The authors report that the precise conditions of this digestion step were tuned to liberate intact nuclei from endothelial, mural, and perivascular cells without compromising their quality, a balance that had defeated earlier attempts to profile vascular cells from frozen tissue.

After extraction, the protocol employs fluorescence-activated cell sorting, or FACS, to ensure that only high-purity nuclei proceed to sequencing. This sorting step is what makes the method compatible with the droplet-based platforms that dominate the field, including the widely used 10x Genomics single cell 3′ gene expression assay and the 10x multiome platform, which captures both gene expression and chromatin accessibility from the same nucleus. In parallel with the vascular nuclei, the protocol recovers parenchymal nuclei from the density gradient, meaning that a single dissection can yield both a vascular and a non-vascular view of the same tissue sample. Both fractions are purified by FACS, ensuring compatibility with single-nucleus RNA sequencing, single-nucleus ATAC sequencing, and combined multiomic workflows.

The practical demands of the protocol are modest by the standards of modern genomics. The authors estimate that the full workflow, from homogenized tissue to sorted nuclei ready for sequencing, requires approximately four to five hours to complete. Researchers with training in single-cell techniques and flow cytometry should be able to carry it out, which lowers the barrier to entry considerably compared with bespoke methods that demand specialized instrumentation or rare expertise. The protocol also accommodates both fresh and frozen tissue, a crucial feature for human research, where most brain samples arrive from brain banks as frozen material. Detailed guidance is provided for interpreting quality-control checkpoints along the way, including the use of Agilent BioAnalyzer traces during downstream library preparation, so that laboratories can diagnose and correct problems before committing precious samples to sequencing.

The significance of the method is best understood through the discoveries it has already enabled. VINE-seq formed the technical backbone of a human brain vascular atlas published in Nature in 2022, which revealed diverse mediators of Alzheimer’s disease risk expressed in specific vascular cell populations. Its multiome successor, MultiVINE-seq, extended the approach to paired measurements of gene expression and chromatin accessibility, and underpinned a 2025 study in Neuron demonstrating that human brain vascular multi-omics can elucidate disease-risk associations that transcriptomics alone cannot resolve. By publishing the complete protocol, the authors are effectively handing the wider research community the keys to a molecular map of the cerebrovasculature that was previously accessible only to a handful of specialist laboratories.

The broader implications reach into nearly every corner of neuroscience and neurology. The blood-brain barrier, formed by tightly joined endothelial cells and supported by pericytes and astrocyte endfeet, controls what enters the brain from the bloodstream and is implicated in systemic infection, inflammation, and the failure of countless drug candidates to reach their targets. Perivascular macrophages, which the protocol captures alongside endothelial and mural cells, are increasingly recognized as immunological sentinels with roles in neurodegeneration. Recent work has linked glycocalyx dysregulation to blood-brain barrier failure in aging, identified angiopoietin signaling as a central axis of amyloid-driven vascular dysfunction, and shown that depletion of the RNA-binding protein TDP-43 in endothelial cells disrupts core barrier pathways in neurodegeneration. Each of these lines of inquiry depends on the ability to molecularly profile vascular cells from real human tissue, precisely what VINE-seq and MultiVINE-seq make routine.

The authors have also made the underlying data openly available, with raw sequencing data for VINE-seq deposited in the NCBI Gene Expression Omnibus under accession GSE163577 and MultiVINE-seq data available in the Sequence Read Archive under BioProject PRJNA1182356. For a field that has spent more than half a century developing methods to isolate brain microvessels, dating back to pioneering capillary isolation work in the 1970s, the arrival of a standardized, hours-long workflow that feeds directly into single-nucleus and multiome sequencing marks a genuine inflection point. As laboratories around the world adopt the protocol, the vascular blind spot in brain atlases is likely to shrink rapidly, and with it, the gaps in our understanding of how the brain’s plumbing shapes neurological health, aging, and disease.

Subject of Research: Single-nucleus and multiome sequencing protocols for profiling cells of the brain vasculature

Article Title: VINE-seq and MultiVINE-seq for single-nucleus and multiome profiling of the brain vasculature

Article References: Oberhauser, J., Ding, B., Reid, M. M., Xie, W. H., & Yang, A. C. (2026). VINE-seq and MultiVINE-seq for single-nucleus and multiome profiling of the brain vasculature. Nature Protocols. https://doi.org/10.1038/s41596-026-01434-x

Image Credits: AI Generated

DOI: 10.1038/s41596-026-01434-x

Keywords: VINE-seq, MultiVINE-seq, brain vasculature, single-nucleus sequencing, blood-brain barrier, endothelial cells, pericytes, perivascular macrophages, FACS, Alzheimer's disease, multiome profiling, Nature Protocols