Optimisation of Xenium automated in situ sequencing for PAXgene-fixed tissue samples

This study presents an optimized workflow using a novel pepsin-based digestion protocol (XTO) to successfully adapt the Xenium automated in situ sequencing platform for PAXgene-fixed tissues, demonstrating that these samples can achieve spatial RNA transcript detection comparable to or superior to standard FFPE samples while preserving tissue morphology.

The Gist

Imagine you are trying to read a library of books, but the books are locked inside a vault. For a long time, scientists have had two main keys to open this vault: one for "fresh frozen" books and another for "formalin-fixed" books. But there's a third type of book, preserved in a special chemical called PAXgene, that has been largely ignored because the existing keys didn't fit. This special chemical is great for keeping the DNA and RNA (the "text" of the book) safe without gluing the pages together, but it makes it hard to read the words inside when using modern, high-tech scanners.

This paper is about finding a new key to unlock those PAXgene books so they can be read by a powerful new scanner called Xenium. Think of Xenium as a super-fast, automated robot librarian that can map exactly where every word (gene) is located on a specific page (tissue sample).

Here is how the scientists solved the puzzle:

• The Problem: The PAXgene vault was too tight. The "pages" (tissue cells) were so packed that the robot librarian couldn't reach the words to read them.
• The Solution: The team created a new recipe called XTO (Xenium Tissue Optimisation). They treated the tissue samples with a digestive enzyme called pepsin.
• The Analogy: Imagine the tissue is a dense, sticky ball of clay. To read the words inside, you need to soften the clay just enough to pull the pages apart without turning the whole book into mush. The scientists found that pepsin acts like a gentle, precise solvent. It "digests" just the right amount of the sticky material to make the RNA (the words) accessible.
• The Balancing Act: They discovered that timing is everything. If you let the pepsin work for too short a time, the words are still hidden. If you let it work too long, the book falls apart, and you lose the picture of what the tissue looked like. They had to find the "Goldilocks" zone—just enough digestion to see the genes clearly, but not so much that the tissue structure is ruined.
• The Result: When they got the timing right, the PAXgene samples didn't just work; they performed as well as, or even better than, the standard formalin-fixed samples. The robot librarian could find more words in the PAXgene books than in the traditional ones.

In short: The paper proves that with a new, carefully timed "digestion" step, we can now use advanced gene-mapping technology on a type of tissue sample (PAXgene) that was previously difficult to use. This opens the door for scientists to use this technology on both old archives and new collections of these specific samples, expanding the library of data they can study.

Technical Summary

Technical Summary: Optimisation of Xenium Automated In Situ Sequencing for PAXgene-Fixed Tissue Samples

The Problem
While spatial transcriptomics has revolutionized the study of gene expression within tissue contexts, its application has largely been restricted to fresh-frozen and formalin-fixed paraffin-embedded (FFPE) samples. The use of non-standard preservation methods remains underexplored. Specifically, PAXgene fixation, which preserves DNA and RNA without the crosslinking typical of formalin, presents unique challenges for high-quality spatial transcriptomics. Although advantageous for genomic studies, adapting PAXgene-fixed paraffin-embedded (PFPE) tissues to automated in situ sequencing platforms like Xenium requires overcoming significant barriers in RNA accessibility and tissue integrity.

Methodology
To address these challenges, the authors developed a specialized workflow termed the Xenium Tissue Optimisation (XTO) protocol. This approach was applied to PFPE tissues derived from multiple mouse and human samples. The core of the methodology involved a systematic evaluation of permeabilization conditions to determine the most effective means of exposing RNA for sequencing. A critical variable in this optimization was the duration of pepsin digestion. The researchers tested various digestion times to identify conditions that would maximize RNA accessibility while preserving the structural integrity of the tissue and the quality of morphological staining.

Key Contributions and Results
The study demonstrates that pepsin digestion is an effective mechanism for enhancing RNA accessibility in PFPE samples. The XTO protocol revealed that digestion times are tunable, allowing for tissue-specific optimization. However, the authors note a necessary trade-off: maximizing transcript detection rates often requires conditions that may impact tissue integrity or morphological staining, necessitating a compromise based on specific experimental goals.

Under these optimized conditions, the results indicate that PFPE samples can achieve spatial RNA transcript detection rates that are comparable to, or in some cases superior to, those obtained from standard FFPE samples.

Significance
This work provides a concrete framework for adapting automated in situ sequencing technologies to PFPE tissues. By establishing a viable workflow for PAXgene-fixed samples, the study broadens the applicability of spatial transcriptomics. This expansion is significant for both archival collections and prospective sample gathering, offering researchers a new avenue to leverage PAXgene-fixed materials for spatial gene expression analysis without sacrificing data quality.