A Workflow for Spatial Transcriptomic Analysis from Intra-operative Human Skeletal Muscle Biopsies

This study establishes the feasibility of applying high-resolution spatial transcriptomics to intra-operative human skeletal muscle biopsies to define fiber-type-specific gene expression signatures associated with neuromuscular degeneration and regeneration following peripheral nerve injury.

The Gist

Imagine your body's muscles are like a massive, bustling city. For years, scientists have studied this city by taking a scoop of the whole neighborhood, grinding it up into a smoothie, and analyzing the mixture. This tells them what ingredients are in the city (the genes), but it completely destroys the map of where those ingredients are located. You lose the ability to see if a bakery is next to a park or if a factory is isolated in a corner.

This paper is about building a high-definition GPS for the human muscle city that doesn't require grinding it up.

Here is the story of how they did it, using simple analogies:

1. The Problem: The "Smoothie" vs. The "Map"

When a nerve is injured (like in a car accident), the muscles it controls start to shrink and weaken. Doctors want to know: Can this muscle recover if we fix the nerve? To answer this, they need to look at the muscle's "instruction manual" (its genes).

Previously, scientists had to chop the muscle up to read the manual. This was like trying to understand a city's layout by drinking a smoothie made of its bricks. You know there are bricks, but you don't know if they form a wall, a road, or a roof.

2. The Solution: The "Super-Microscope"

The researchers used a new technology called Spatial Transcriptomics (specifically the 10x Genomics Visium HD). Think of this as a magical camera that can take a photo of a muscle slice and, at the same time, read the "ID card" of every single cell in that photo without moving them.

• The Resolution: They didn't just look at the whole muscle; they looked at it in tiny 8-micrometer squares. That's like zooming in so far you can see the individual rooms inside a house, not just the house itself.
• The Source: They took these "photos" from real human patients during routine surgery. This is like getting a live update from the city while it's still running, rather than studying an old, abandoned map.

3. What They Found: The Muscle's "Neighborhoods"

Using this new map, they discovered that muscles are much more organized than we thought.

• The Fiber Types: Muscles have different types of "workers" (fibers). Some are slow and steady (Type I, like marathon runners), and some are fast and explosive (Type II, like sprinters). The researchers could see exactly where the marathon runners lived and where the sprinters lived, side-by-side in the same muscle.
• The "Core" vs. The "Rim": Even inside a single muscle fiber, they found different neighborhoods. The center of the fiber (the core) had one set of instructions, while the outer edge (the rim) had a different set. It's like finding that the kitchen in a house has a different set of tools than the living room, even though they are in the same building.
• The "Neuromuscular Junction" (The Power Switch): This is the most exciting part. The neuromuscular junction is where the nerve talks to the muscle. It's the "power switch." The researchers found that the genes responsible for this switch aren't scattered randomly. They are clustered in tight, specific "hotspots," like a group of people huddled around a campfire. They could actually see these clusters on the map.

4. Why This Matters: The "Crystal Ball" for Surgery

Why does this matter for a regular person?

Imagine a patient has a severed nerve. The doctor has to decide: Should we try to reconnect the nerve now, or wait? Will the muscle be able to wake up again?

Right now, doctors are guessing. But with this new "GPS map," they might be able to look at a tiny piece of the patient's muscle and say:

"Look at this map. The 'power switches' (neuromuscular junctions) are still intact and clustered together. This muscle is ready to be reconnected!"

Or conversely:

"The map shows the 'power switches' are scattered and broken. This muscle might be too damaged to recover, so we need a different plan."

The Bottom Line

This paper is a "proof of concept." It's like showing that a new, super-powerful drone can fly over a city, take high-res photos, and read street signs simultaneously.

They proved it works on human muscle taken during surgery. While they only looked at one person in this study, they have built the workflow (the instructions) for how to do this in the future. This paves the way for a future where doctors can use these high-tech maps to make better decisions about nerve repairs, helping more people regain movement after injuries.

In short: They stopped grinding up the muscle to study it. Instead, they built a detailed, living map that shows exactly where the muscle's "instructions" are written, opening the door to better cures for nerve injuries.

Technical Summary

1. Problem Statement

Peripheral nerve injuries often lead to muscle denervation, atrophy, and functional loss. While nerve repair or transfer can restore axonal contact, functional recovery is inconsistent due to reinnervation failure. Current methods to assess reinnervation competence (e.g., motor-endplate morphometry) are labor-intensive and provide only a partial proxy for the broader transcriptional and microenvironmental changes governing recovery.

• Limitation of Existing Techniques: Conventional single-cell RNA sequencing (scRNA-seq) requires tissue dissociation, destroying spatial context. Single-nucleus RNA-seq (snRNA-seq) retains architecture but loses extranuclear RNA information.
• The Gap: There is a critical need for high-resolution, spatially resolved gene expression data from human skeletal muscle obtained in clinically relevant, intra-operative settings to identify molecular signatures of successful reinnervation.

2. Methodology

The study established a workflow to apply 10x Genomics Visium HD (High-Definition) spatial transcriptomics to human muscle biopsies.

• Sample Collection: A trapezius muscle biopsy was obtained intra-operatively from a 24-year-old male undergoing orthopedic surgery for traumatic brachial plexus injury (5 months post-injury). The muscle was confirmed to be normally innervated (control) relative to the injured nerves.
• Sample Preparation:
  • Biopsies were flash-frozen in liquid nitrogen-cooled isopentane and stored at -80°C.
  • Sections were cut at 10 µm thickness.
  • RNA Integrity Number (RIN) was assessed (RIN = 7.7), confirming high-quality RNA.
  • Sections were stained with Hematoxylin and Eosin (H&E) for histological alignment.
• Spatial Transcriptomics Workflow:
  • Platform: 10x Genomics Visium HD (8 µm resolution bins) using the whole human transcriptome probe panel.
  • Library Prep: Performed via CytAssist-Enabled Probe Release & Capture.
  • Sequencing: Whole-transcriptome sequencing performed at the UCI Genomics Research and Technology Hub.
• Data Analysis Pipeline:
  • Dual-Approach: Data was analyzed using both Seurat v5 (R-based computational framework) and Loupe Browser (10x Genomics visualization tool).
  • Quality Control (QC): Applied thresholds for Unique Molecular Identifiers (UMIs), gene counts, and mitochondrial gene percentage to filter low-quality bins.
  • Dimensionality Reduction: Used Principal Component Analysis (PCA) and Uniform Manifold Approximation and Projection (UMAP) for clustering.
  • Fiber Typing: Assigned spatial bins to fiber types (Type I, Type II, IIA, IIX) based on composite scores of canonical marker genes (e.g., MYH7, MYH1, TNNT1).
  • Spatial Statistics: Performed Moran's I spatial autocorrelation analysis and Local Indicators of Spatial Association (LISA) to identify "High-High" (HH) hotspots of gene co-expression, specifically for Neuromuscular Junction (NMJ) genes.

3. Key Contributions

• Feasibility Demonstration: Proved that high-resolution spatial transcriptomics can be successfully applied to intra-operative human muscle biopsies, overcoming challenges related to tissue handling and RNA integrity in a clinical setting.
• Subcellular Resolution: Achieved 8 µm resolution, allowing for the mapping of gene expression not just to cell types, but to subcellular domains within multinucleated muscle fibers (e.g., distinguishing the central contractile core from the peripheral sarcolemma).
• Methodological Framework: Established a complementary workflow using Seurat for robust statistical clustering and Loupe Browser for high-resolution, anatomy-aware visualization and QC.
• Human-Specific Baseline: Generated the first spatially resolved baseline gene expression signatures for human trapezius muscle, including fiber-type composition and NMJ organization.

4. Key Results

• Data Quality: The dataset showed robust RNA capture (median 99.6 UMIs per bin, 71.9 genes detected) with high sequencing saturation (63%) and low mitochondrial content (5%), indicating well-preserved tissue.
• Cellular Heterogeneity: The analysis resolved 21 distinct transcriptional populations, including:
  • Myofibers: Type I (slow, 12.2%) and Type II (fast, 33.4%) fibers.
  • Supporting Cells: Fibro-adipogenic progenitors (FAPs), immune cells, vascular/endothelial cells, and nerve cells.
• Subcellular Compartmentalization:
  • Type I Fibers: Cores enriched with slow-contractile genes (TNNC1, TNNT1); rims enriched with calcium cycling modulators (MYLK3, PPP1R1C).
  • Type II Fibers: Cores enriched with fast-contractile genes (MYH1, MYH2); rims enriched with calcium-handling markers (PVALB, CALML6).
• Fiber Type Composition: The trapezius sample was predominantly Type II fibers, with a specific enrichment for Type IIA (oxidative-glycolytic) over Type IIX (glycolytic).
• NMJ Spatial Clustering:
  • NMJ-associated genes (e.g., CHRNA1, CHRNB1, LRP4, AGRN, MUSK) showed significant non-random spatial clustering (High Moran's I Z-scores).
  • Hotspot Analysis: Buffered "High-High" (HH) analysis confirmed that NMJ transcripts co-localize in discrete microdomains within the muscle fibers, consistent with subsynaptic localization.
  • Genes like CHRNE and COL13A1 did not show significant clustering, highlighting gene-specific spatial regulation.

5. Significance and Future Implications

• Clinical Translation: This workflow provides a framework for identifying biomarkers that predict successful reinnervation. By mapping molecular signatures in denervated vs. reinnervated muscle, surgeons could potentially optimize the timing of nerve transfer surgeries.
• Disease Mechanism: The ability to resolve subcellular domains (core vs. rim) and NMJ microdomains offers new insights into neuromuscular diseases where local mRNA translation and synaptic organization are critical.
• Discovery Platform: While currently limited by cost and processing time for large cohorts, this method serves as a discovery engine. Identified spatial biomarkers can eventually be translated into scalable, routine clinical assays.
• Human vs. Model Organisms: The study highlights potential human-specific fiber-type programs and isoform shifts that may differ from mouse models, emphasizing the necessity of human tissue studies for translational research.

In conclusion, the paper successfully validates a high-resolution spatial transcriptomic workflow for human muscle, revealing intricate subcellular gene expression patterns and NMJ organization that were previously inaccessible with dissociative methods.