Single-nucleus RNA sequencing (snRNA-seq) has become the method of choice for researchers working with tissues that resist enzymatic dissociation. While single-cell RNA-seq (scRNA-seq) requires intact, viable cells, snRNA-seq isolates nuclei directly — bypassing the problems that make certain tissue types notoriously difficult to process. This guide walks through the specific challenges posed by brain, skeletal muscle, adipose tissue, heart, kidney, and fibrotic samples, and explains how snRNA-seq addresses each one.

Tissue dissociation is not a one-size-fits-all process. Some tissues break apart easily in enzyme cocktails; others hold together stubbornly no matter how long the incubation. The difference often comes down to extracellular matrix composition, cell size and shape, lipid content, and the presence of multinucleated or syncytial structures. Understanding these tissue-specific obstacles helps researchers decide early whether snRNA-seq is the right path — before precious samples are consumed on failed dissociation attempts.

Why Some Tissues Resist Dissociation

Enzymatic tissue dissociation works by digesting the extracellular matrix (ECM) and cell-cell junctions that hold cells in place. In theory, this releases individual cells into suspension. In practice, several features of difficult tissues interfere with this process.

Dense or crosslinked ECM. Tissues rich in collagen, elastin, and other structural proteins — such as heart, kidney, and fibrotic organs — resist enzymatic digestion. Even prolonged incubation with collagenase or liberase may yield incomplete dissociation and a high proportion of damaged cells [1].

Large or fragile cell bodies. Neurons extend processes that shear during mechanical trituration. Mature cardiomyocytes can exceed 100 µm in length and are prone to rupture. Adipocytes are filled with a single large lipid droplet that makes them buoyant and fragile [2].

Multinucleation. Skeletal muscle fibers and some cardiomyocytes contain hundreds of nuclei within a shared cytoplasm. Enzymatic dissociation cannot separate individual nuclei from these syncytial structures — only nuclei isolation can [3].

Stress-induced transcriptional artifacts. Prolonged enzymatic incubation at 37°C activates stress-response and apoptotic gene programs. This is well documented in kidney and heart tissues, where dissociation-induced expression changes can confound biological interpretation [4].

snRNA-seq sidesteps these problems by isolating nuclei through mechanical homogenization and detergent-based lysis rather than enzymatic digestion. The harsh but brief lysis conditions strip away cytoplasm, ECM fragments, and lipid droplets, while preserving intact nuclei for downstream processing with single-nucleus RNA sequencing platforms.

Figure 1: Six tissue types that present distinct challenges for enzymatic dissociation, and why single-nucleus RNA sequencing is often the preferred approach for each.

Brain and Neuronal Samples

Brain tissue was one of the first difficult sample types to drive widespread adoption of snRNA-seq. Adult neurons are large, highly ramified cells whose processes can extend millimeters from the soma. Mechanical or enzymatic dissociation shreds these processes and often destroys the cell body entirely.

The dissociation problem. Neuronal processes create entanglement during tissue trituration. Even when some intact cell bodies are recovered, scRNA-seq from adult brain consistently underrepresents neurons and overrepresents smaller, less fragile glial cells — a bias that distorts cell-type proportions [5].

How snRNA-seq helps. Nuclei isolation protocols for brain tissue use Dounce homogenization in a detergent-containing lysis buffer. The detergent dissolves cytoplasmic and plasma membranes, while the gentle mechanical force releases nuclei from the tissue matrix. Protocols optimized for adult mouse and human brain are well established, including the official 10x Genomics nuclei isolation protocol and detailed step-by-step guides in Current Protocols [6].

What to expect. Brain snRNA-seq captures diverse neuronal subtypes, astrocytes, oligodendrocytes, microglia, and vascular cells. The proportion of neuronal nuclei to glial nuclei more closely matches histological expectations than scRNA-seq from the same tissue. Researchers should note that snRNA-seq primarily captures nuclear and nascent transcripts, so certain cytoplasmic mRNAs — particularly those with long half-lives in neuronal processes — may appear depleted relative to whole-cell methods.

Skeletal Muscle and Multinucleated Fibers

Skeletal muscle presents a structural challenge that enzymatic dissociation cannot solve: the syncytial nature of muscle fibers.

The multinucleation barrier. Each mature skeletal muscle fiber is a single large cytoplasmic compartment containing hundreds of myonuclei positioned along its length. The fiber is encased in a tough basal lamina rich in collagen and laminin. Enzymatic digestion may strip away the basal lamina, but it cannot separate individual myonuclei — the entire fiber remains a single multinucleated entity that is far too large for microfluidic droplet-based platforms [3].

Figure 2: Comparison of enzymatic dissociation (left) versus nuclei isolation (right) for multinucleated skeletal muscle fibers. Only nuclei isolation releases individual myonuclei for single-cell-resolution transcriptomic analysis.

snRNA-seq as the only viable option. By mechanically disrupting the muscle fiber and lysing the cytoplasmic membrane, nuclei isolation releases individual myonuclei for capture. A key study by Petrany et al. (2020) demonstrated that snRNA-seq from skeletal muscle captures transcriptional heterogeneity across individual myonuclei within the same fiber — information that is completely inaccessible by any whole-cell method [3].

Cell-type recovery. Beyond myonuclei, skeletal muscle snRNA-seq captures fibro-adipogenic progenitors (FAPs), satellite cells, endothelial cells, and immune cells. Notably, FAP recovery in snRNA-seq is comparable to or better than scRNA-seq, likely because these cells are also embedded in the fibrous matrix and are lost during enzymatic dissociation [7].

Practical considerations. Fibrous muscle tissue requires more aggressive homogenization than brain. Pre-mincing the tissue into small fragments before Dounce homogenization improves yield. Expect more debris in the nuclei suspension compared to softer tissues — additional filtration or density gradient steps may be needed before loading onto a microfluidic platform.

Adipose Tissue and Lipid Interference

Adipose tissue creates a different kind of problem: lipid overwhelms everything.

The buoyancy problem. Mature white adipocytes contain a single large lipid droplet that occupies >90% of the cell volume, squeezing the nucleus and cytoplasm to the periphery. During cell suspension preparation, these lipid-filled cells float to the surface, resist pelleting by centrifugation, and break apart easily. Even when adipocytes are successfully captured, the lipid content can interfere with microfluidic droplet formation and reverse transcription [8].

Lipid droplets and debris. Adipose tissue homogenization releases a flood of lipid droplets that coat tubing, clog microfluidic channels, and create a scum layer that obscures nuclei during counting and sorting. Standard scRNA-seq protocols designed for hematopoietic or epithelial cells simply do not accommodate this level of lipid contamination.

snRNA-seq bypasses lipid. Nuclei isolation strips away the lipid droplet along with the cytoplasm. The resulting nuclear pellet is free of buoyancy artifacts and compatible with standard microfluidic workflows. A 2025 study by So et al. established a robust snRNA-seq protocol for adipose tissue, demonstrating that adipocyte nuclei cluster distinctly from stromal vascular fraction nuclei and retain cell-type-specific transcriptional signatures [8].

What is captured. Adipose snRNA-seq captures nuclei from mature adipocytes, preadipocytes, endothelial cells, macrophages, and other immune and stromal cells. The adipocyte nuclear proportion is substantially higher than what scRNA-seq typically recovers from the same depots, providing a more accurate picture of tissue composition.

Cardiac Muscle and Dense Matrix

The heart is one of the most challenging tissues for single-cell transcriptomics, combining dense ECM, large cell size, and binucleation.

Why cardiac dissociation fails. Adult human cardiomyocytes are binucleated cells approximately 100-150 µm long, embedded in a dense network of collagen and elastin fibers. Enzymatic dissociation protocols require Langendorff perfusion or extended incubation periods that activate stress-response genes in cardiomyocytes. The resulting cell suspension is often dominated by smaller non-myocyte populations, with cardiomyocytes either destroyed during processing or too large for microfluidic capture [9].

Nuclei isolation for cardiac tissue. Dedicated cardiac nuclei isolation protocols have been developed specifically for snRNA-seq. The Current Protocols method from Litviňuková et al. (2022) uses detergent-based lysis with careful optimization of detergent concentration and incubation time to avoid nuclear membrane damage [9]. The SoNIC method (Single-Nuclei Isolation from Cardiac tissue) further refines this approach for postmortem human heart samples [10].

Landmark cardiac snRNA-seq studies. A major 2022 study by Chaffin et al. published in Nature profiled left ventricular samples from 44 individuals — including patients with dilated cardiomyopathy, hypertrophic cardiomyopathy, and non-failing hearts — using snRNA-seq. The study identified disease-specific cell-state changes and a population of activated fibroblasts in cardiomyopathy hearts that would have been invisible with bulk RNA-seq and technically impossible with scRNA-seq [11].

Tissue quality matters. Cardiac nuclei isolation is sensitive to tissue handling. Ischemic time between tissue collection and freezing should be minimized. RNAlater-treated samples yield lower-quality nuclei than snap-frozen tissue. Expect higher debris levels from cardiac tissue than from brain or adipose — rigorous filtration and debris-removal QC steps are essential.

Kidney and ECM-Rich Architecture

The kidney's dense tubular basement membranes and complex ECM architecture make it a classic "difficult tissue" for single-cell workflows.

Stress artifacts from dissociation. A systematic comparison of kidney dissociation protocols by Denisenko et al. (2020) demonstrated that even optimized enzymatic protocols induce substantial transcriptional stress. Genes related to apoptosis, heat shock, and immediate-early response are activated during the 30-60 minute incubation typical of kidney dissociation — artifacts that can masquerade as biological signals in scRNA-seq data [4].

snRNA-seq for kidney. Nuclei isolation from kidney tissue uses mechanical disruption — typically Dounce homogenization or gentleMACS processing — in a detergent buffer. This avoids the prolonged 37°C incubation that drives stress artifacts. snRNA-seq from kidney captures podocytes, proximal and distal tubular epithelial cells, collecting duct cells, mesangial cells, endothelial cells, and immune cells [12].

Biopsy compatibility. Kidney research often relies on needle biopsy specimens where tissue is extremely limited. snRNA-seq is compatible with small input amounts — nuclei can be isolated from as little as 5-10 mg of tissue, making it feasible for biopsy-based studies. A 2024 abstract reported successful snRNA-seq from human kidney biopsy tissue, expanding the range of samples available for transcriptomic analysis [12].

Debris management. Kidney tissue generates significant debris during homogenization, including basement membrane fragments and tubular casts. Density gradient centrifugation (e.g., iodixanol or sucrose cushion) is strongly recommended to separate clean nuclei from debris before library preparation. Debris-removal algorithms such as DIEM can further improve data quality during computational processing [13].

Fibrotic and Scarred Samples

Fibrotic tissue represents perhaps the most extreme challenge for tissue dissociation — and the strongest case for snRNA-seq.

ECM deposition blocks dissociation. Fibrosis is characterized by excessive deposition of collagen, fibronectin, and other ECM proteins. This dense, crosslinked matrix physically blocks enzyme access to cells embedded within it. In advanced fibrosis — whether hepatic, pulmonary, renal, or cardiac — enzymatic dissociation protocols that work on healthy tissue often fail completely [14].

snRNA-seq as the only single-cell option. For fibrotic specimens, nuclei isolation is frequently the only route to single-cell-resolution transcriptomic data. The mechanical and detergent-based approach breaks through the ECM barrier that enzymatic digestion cannot penetrate. A protocol from Ramachandran et al. demonstrates nuclei isolation from snap-frozen fibrotic human liver, producing high-quality snRNA-seq data that reveals fibroblast activation states and immune cell dynamics within the fibrotic niche [14].

Higher debris expectations. Fibrotic samples release more ECM fragments during homogenization than healthy tissue. This translates to more debris in the nuclei suspension and potentially higher background signal in the final data. Researchers should budget for additional washing steps, more stringent filtration, and post-sequencing debris-filtering algorithms such as QClus, which was specifically developed for debris-heavy snRNA-seq data [15].

Cross-organ applicability. The principles that make snRNA-seq effective for fibrotic liver apply equally to fibrotic lung (idiopathic pulmonary fibrosis), fibrotic kidney (CKD with interstitial fibrosis), and fibrotic heart (post-infarction scar). In each case, the ECM barrier is the primary obstacle, and nuclei isolation is the most reliable solution.

Nuclei Quality Across Tissue Types

Nuclei quality is not uniform across tissues. Each tissue type produces nuclei with characteristic features that affect downstream data quality.

What to monitor. Key nuclei QC metrics include nuclear membrane integrity (assessed by DAPI or trypan blue staining), the absence of cytoplasmic tags, and the proportion of debris versus intact nuclei in the final suspension. A minimum of 70-80% intact nuclei is generally acceptable, though fibrotic and cardiac samples may fall closer to 60-70% and still yield usable data.

Figure 3: Representative nuclei quality across three tissue types. Brain nuclei are typically the cleanest, while cardiac and fibrotic samples require more stringent filtration and debris-removal steps.

Tissue-specific expectations. Brain nuclei suspensions are typically the cleanest, with well-defined round nuclei and minimal debris. Cardiac and skeletal muscle nuclei are more variable in size and shape and often carry adherent debris. Adipose nuclei are small and may require careful counting with a fluorescent nuclear stain. Kidney and fibrotic nuclei suspensions contain more background particles; density gradient purification is strongly recommended for these sample types.

Post-sequencing QC. After library preparation and sequencing, standard scRNA-seq QC metrics apply: number of genes detected per nucleus, proportion of mitochondrial reads (should be low in snRNA-seq since mitochondria are cytoplasmic), and doublet rates. The DIEM algorithm can computationally identify and remove droplets dominated by debris-derived ambient RNA rather than genuine nuclear transcripts [13].

When scRNA-seq Still Works

snRNA-seq is not always necessary. Understanding when scRNA-seq remains the better choice helps avoid over-engineering a study.

Fresh tissue with easy dissociation. Tissues such as spleen, lymph node, bone marrow, and peripheral blood dissociate readily with minimal enzymatic treatment. scRNA-seq from these samples captures both nuclear and cytoplasmic transcripts, including mitochondrial genes and non-coding RNAs that are depleted in snRNA-seq.

Immune-cell-focused studies. scRNA-seq enriches for immune cells relative to snRNA-seq, as demonstrated by matched comparisons in colon and liver tissue. Immune cells are less adherent and survive dissociation better, while adherent cell types (epithelial, hepatocyte, fibroblast) are selectively lost during scRNA-seq processing. If the primary research question concerns immune cell populations, scRNA-seq from fresh tissue may actually provide better immune cell recovery than snRNA-seq [5].

When cytoplasmic RNA is essential. snRNA-seq captures predominantly nuclear and nascent transcripts, which means certain cytoplasmic RNA species — including some long non-coding RNAs, mature mRNAs sequestered in processing bodies, and transcripts with extensive cytoplasmic polyadenylation — are underrepresented. For questions that require full-length cytoplasmic transcript coverage, scRNA-seq from fresh tissue remains the gold standard.

Feasibility Checklist for Difficult Tissues

Before committing samples to snRNA-seq, run through these tissue-specific feasibility questions.

1. Has your tissue type been successfully processed by snRNA-seq in published literature? If not, a pilot nuclei isolation on a small tissue aliquot is recommended before committing the full sample set.
2. Is your tissue frozen, archived, or difficult to dissociate? If yes to any of these, snRNA-seq is the appropriate starting point. Contact our team to discuss sample feasibility and receive protocol recommendations for your tissue type.
3. Do you expect your target cell population to be adherent, large, fragile, or embedded in dense ECM? Neurons, cardiomyocytes, adipocytes, myocytes, fibroblasts, podocytes, and hepatocytes are all better captured by snRNA-seq than scRNA-seq.
4. Is your sample fibrotic, calcified, or from an aged donor? These features amplify dissociation difficulty. Plan for snRNA-seq and budget for additional QC steps.
5. Do you have enough tissue for a pilot nuclei isolation? A small-scale test run (5-10 mg of tissue) can verify nuclei yield and quality before proceeding with the full experiment.
6. Is your research question compatible with nuclear transcriptomes? If your question requires cytoplasmic transcripts, mitochondrial genes, or full-length mRNA coverage, consider whether matched bulk RNA-seq or targeted assays can supplement the snRNA-seq data.

For guidance on tissue preservation, storage, and shipping before nuclei isolation, see our article on frozen tissue handling for snRNA-seq. Our snRNA sequencing services include sample feasibility review, nuclei isolation optimized for your tissue type, library preparation, sequencing, and bioinformatics analysis — from raw data through cell-type annotation, differential expression, and pathway enrichment.

FAQ

Q: Can snRNA-seq work on FFPE tissue?

A: Standard snRNA-seq protocols are designed for fresh or frozen tissue. FFPE tissue requires specialized protocols for RNA extraction from crosslinked samples. For spatial transcriptomics from FFPE tissue, FFPE spatial transcriptomics services may be more appropriate. Contact us to discuss your specific sample type.

Q: How much tissue do I need for snRNA-seq from difficult samples?

A: As little as 10-30 mg of tissue can yield sufficient nuclei for snRNA-seq, though fibrotic or highly fibrous tissues may require more starting material. Exact amounts depend on tissue cellularity — discuss your sample with our team for tissue-specific guidance.

Q: Will I lose immune cell information by using snRNA-seq?

A: Immune cells are proportionally underrepresented in snRNA-seq compared to scRNA-seq because snRNA-seq enriches for adherent structural cells. However, immune cell types are still detected. If immune populations are your primary focus and fresh tissue is available, scRNA-seq may be preferable. For integrated studies, matched snRNA-seq plus a focused immune profiling approach can provide comprehensive coverage.

Q: Can I process multiple tissue types in the same snRNA-seq batch?

A: Yes, but nuclei isolation protocols may differ between tissue types. Our team can optimize isolation conditions for each tissue type in a multi-tissue study. Discuss your project design with us to plan sample processing and minimize batch effects.

Q: What are the most common causes of snRNA-seq failure in difficult tissues?

A: The top three causes are: (1) insufficient tissue amount or poor tissue quality (prolonged warm ischemia); (2) nuclei damaged during isolation due to over-homogenization or inappropriate detergent concentration; and (3) excessive debris overwhelming the nuclei suspension due to skipped or inadequate filtration/density gradient steps. Following tissue-optimized protocols and including QC checkpoints at each stage greatly reduces failure rates.

References

1. Litviňuková M, et al. "Cells of the adult human heart." Nature. 2020;588:466-472. doi:10.1038/s41586-020-2797-4
2. Nguyen QH, et al. "Profiling human breast epithelial cells using single cell RNA sequencing identifies cell diversity." Nature Communications. 2018;9:2028. Note: Demonstrates cell-size bias in scRNA-seq.
3. Petrany MJ, et al. "Single-nucleus RNA-seq identifies transcriptional heterogeneity in multinucleated skeletal myofibers." Nature Communications. 2020;11:6374. doi:10.1038/s41467-020-20063-w
4. Denisenko E, et al. "Systematic assessment of tissue dissociation and storage biases in single-cell and single-nucleus RNA-seq workflows." Genome Biology. 2020;21:130. doi:10.1186/s13059-020-02048-6
5. Oh JM, et al. "Comparison of cell type distribution between single-cell and single-nucleus RNA-sequencing: enrichment of adherent cell types in single-nucleus RNA-sequencing." Experimental & Molecular Medicine. 2023;55:273-283. doi:10.1038/s12276-022-00892-z
6. Bakken TE, et al. "Single-Nucleus RNA-Sequencing in Brain Tissue." Current Protocols. 2023;3:e919. doi:10.1002/cpz1.919
7. Chemello F, et al. "Degenerative and regenerative pathways underlying Duchenne muscular dystrophy revealed by single-nucleus RNA sequencing." STAR Protocols. 2021;2(3):100701. doi:10.1016/j.xpro.2021.100701
8. So J, et al. "Robust single-nucleus RNA sequencing reveals depot-specific cell population dynamics in adipose tissue remodeling during obesity." eLife. 2025;13:RP97981. doi:10.7554/eLife.97981
9. Litviňuková M, et al. "Isolating Nuclei From Frozen Human Heart Tissue for Single-Nucleus RNA Sequencing." Current Protocols. 2022;2:e480. doi:10.1002/cpz1.480
10. Safabakhsh S, et al. "High-quality nuclei isolation from postmortem human heart muscle tissues for single-cell studies." Journal of Molecular and Cellular Cardiology. 2023;179:19-29. doi:10.1016/j.yjmcc.2023.03.009
11. Chaffin M, et al. "Single-nucleus profiling of human dilated and hypertrophic cardiomyopathy." Nature. 2022;608:174-180. doi:10.1038/s41586-022-04817-8
12. Park S, Cho A, et al. "#161 Single-Nucleus RNA-Seq of Human Kidney Biopsy Tissue Reveals Microstructure-Specific Similarities and Differences in Various Glomerulonephritis." Nephrology Dialysis Transplantation. 2024;39(Supplement_1):gfae069-0306-161.
13. Alvarez M, et al. "Enhancing droplet-based single-nucleus RNA-seq resolution using the DIEM computational model." Scientific Reports. 2020;10:10890. doi:10.1038/s41598-020-67513-5
14. Ramachandran P, et al. "Resolving the fibrotic niche of human liver cirrhosis at single-cell level." Nature. 2019;575:512-518. doi:10.1038/s41586-019-1631-3. Note: Protocol extended in PMC9976782 for snap-frozen fibrotic liver nuclei isolation.
15. Schmauch E, Ojanen J, Galani K, Jalkanen J, Harju K, Hollmén M, Kokki H, Gunn J, Halonen J, Hartikainen J, Kiviniemi T, Tavi P, Kaikkonen MU, Kellis M, Linna-Kuosmanen S. "QClus: a droplet filtering algorithm for enhanced snRNA-seq data quality in challenging samples." Nucleic Acids Research. 2024;53(1):gkae1145. doi:10.1093/nar/gkae1145