Introduction
The "pig cell landscape" represents a significant milestone in porcine genomics, offering a high-resolution map of tissue heterogeneity in Sus scrofa. As a primary model for xenotransplantation and human disease, accurate cell typing in pigs is critical for translational research. This correction to the original study provides essential technical updates regarding gene markers for endothelial transition and standardized quality control (QC) protocols for porcine single-cell RNA sequencing (scRNA-seq).
Methodological Refinements: Mitochondrial Quality Control
A significant portion of this correction focuses on the bioinformatic pipeline used to process the atlas. In scRNA-seq, mitochondrial (MT) gene expression is a standard proxy for cell health; high MT levels typically indicate cell stress or lysis. The authors clarify that for the pig cell landscape, cells were retained if they expressed between 200 and 5,000 genes and possessed less than 30% mitochondrial transcripts.
Crucially, the authors identified that the ND1 gene should be excluded from this QC calculation due to high sequence variance in pigs, which could otherwise skew data quality assessments. The finalized list of 12 MT genes (including ATP6, COX1-3, and ND2-6) establishes a reproducible standard for porcine transcriptomics, ensuring that healthy cells are not inadvertently filtered out due to species-specific genetic variance.
Endothelial Heterogeneity and EndMT Identification
The correction refines the identification of Endothelial-to-Mesenchymal Transition (EndMT) cells within adipose tissue. This transition is a key process in vascular remodeling and fibrosis. The authors clarify the following marker associations:
- Primary Markers: The co-expression of VWF and TAGLN serves as the primary identifier for the EndMT population in adipose tissue.
- Secondary Subsets: A smaller fraction of endothelial cells (ECs) co-express PECAM1 and ACTA2.
This distinction is vital for researchers using the pig model to study cardiovascular disease, as it allows for more precise targeting and identification of transitioning vascular cells.
Corrected Visual Evidence
Technical errors in the original Figure 3 legends were addressed to ensure alignment between the transcriptomic data and histological staining. These updates include:
- Figure 3i/j: Confirmation of VWF and TAGLN co-expression in EndMT populations through immunofluorescence.
- Figure 3k: Corrected labeling for FABP4 (fatty acid-binding protein) staining. In the revised version, green arrows correctly point to endothelial cells, while red arrows indicate adipocytes.
Conclusion
While these corrections do not alter the fundamental conclusions of the pig cell landscape, they provide the technical precision necessary for the scientific community to utilize the dataset effectively. The exclusion of the ND1 gene from QC filters highlights the necessity of species-specific adjustments in bioinformatics, ensuring that mitochondrial diversity across different pig breeds does not lead to the loss of high-quality transcriptomic data.
Summary updated