Dataset Information

Highly sensitive, sub-micron spatial host-microbiome transcriptomics in FFPE tissues via iterative hydrogel expansion

ABSTRACT: Spatial transcriptomics (ST) of clinical formalin-fixed paraffin-embedded (FFPE) tissues is fundamentally limited by a trade-off between spatial resolution and transcriptomic sensitivity due to severe molecular crowding and RNA degradation. Here, Ex-spRandom is presented, a highly sensitive platform that integrates random-primed in situ cDNA synthesis with an interpenetrating polymer network (IPN)-based tissue expansion protocol. This physicochemical synergy physically decrowds the dense FFPE matrix, directly converting fragmented host and microbial RNAs into stably anchored cDNA. Consequently, Ex-spRandom translates macro-scale arrays into a sub-micron effective resolution of ~0.9 µm while achieving an unprecedented transcriptomic sensitivity exceeding 40,000 unique genes per sample. This platform successfully rescues non-coding regulatory transcripts and unmasks continuous neurodevelopmental trajectories in the dense embryonic eye. Furthermore, it enables the simultaneous, high-fidelity spatial mapping of host epithelial architectures alongside localized resident microbiomes in the colon. By breaking the resolution-sensitivity barrier without aggressive proteolysis, Ex-spRandom establishes a highly scalable framework for deciphering complex structural and functional microenvironments across vast archival clinical cohorts.

ORGANISM(S): Mus musculus

PROVIDER: GSE324753 | GEO | 2026/08/26

REPOSITORIES: GEO

ACCESS DATA

Dataset's files

Source:

			Action	DRS
		Other

Items per page:

1 - 1 of 1

Similar Datasets

Project description:Understanding gene expression within its spatial context is essential for unravelling biological processes. Laser Capture Microdissection (LCM) has emerged as a transformative technology, enabling targeted isolation of individual cells or regions from tissue sections while preserving spatial context. However, its application to formalin-fixed, paraffin-embedded (FFPE) tissues has been limited by RNA degradation, leaving the vast repository of clinical FFPE samples underutilized. To address this, we introduce LCM-FFPEseq, a novel method combining LCM with the advanced Smart-seq3xpress protocol and FFPE-specific adaptations for spatial transcriptomics of FFPE sections. Unlike traditional protocols requiring thousands of cells to generate high-quality libraries, LCM-FFPEseq achieves high sensitivity, reproducibility, and transcript coverage. With as few as 30 FFPE-embedded K562 cells and Sertoli cells, we detected over 14,000 protein-coding genes per sample, with no substantial gains when a higher number of cells were isolated. Even individual LCM-isolated cells yielded an average of 7,353 or 6,490 protein-coding genes per K562 or Sertoli single cell, respectively. To demonstrate its clinical utility, we applied LCM-FFPEseq to archived testicular FFPE samples from transgender females receiving gender-affirming hormone therapy. Transcriptomic profiling of isolated seminiferous tubules revealed tubular hyalinization to be associated with greater upregulation of extracellular matrix remodelling and inflammatory pathways, alongside stronger downregulation of spermatogenesis-associated pathways. These findings suggest that testicular fibrosis and/or tubular hyalinization may contribute to germ cell loss following inappropriate hormonal exposure. By enabling high-resolution transcriptomics in archived FFPE samples, LCM-FFPEseq unlocks new possibilities for investigating rare cell types, spatial heterogeneity, and therapy-induced tissue remodelling in vast FFPE repositories.

Dataset Information

Highly sensitive, sub-micron spatial host-microbiome transcriptomics in FFPE tissues via iterative hydrogel expansion

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets