Single-Cell RNA Sequencing Outperforms Single-Nucleus RNA Sequencing in Analyzing Pancreatic Cell Diversity and Gene Expression in Goats

Jie Cheng, Tianxi Zhang, Yan Cheng, Kefyalew Gebeyew, Zhiliang Tan, Zhixiong HeInternational Journal of Molecular Sciences 2025 · DOI

TL;DR

This study directly compares single-cell RNA sequencing (scRNA-seq) and single-nucleus RNA sequencing (snRNA-seq) for analyzing the goat pancreas. The findings reveal that scRNA-seq provides a more comprehensive view of pancreatic cell diversity and gene expression, particularly for digestive enzymes, making it the preferred method for this tissue.

What Problem Does This Paper Address?

The pancreas is a vital organ for both endocrine and exocrine functions, playing a critical role in digestion and metabolic regulation. Understanding its cellular composition and gene expression is crucial for research. However, the inherent instability of pancreatic tissue and the challenges associated with obtaining high-quality nucleic acids have made it difficult to determine the optimal single-cell sequencing technology for its study. While both scRNA-seq and snRNA-seq are powerful tools, they have different strengths and limitations, and their comparative efficacy in the goat pancreas remained unexplored. This paper aims to resolve this by systematically comparing these two methods in the context of goat pancreatic research.

Methods

The researchers collected pancreatic tissues from three healthy 10-day-old female Xiangdong black goats. These tissues were processed into two types of suspensions: single-cell suspensions for scRNA-seq and single-nucleus suspensions for snRNA-seq. Both types of samples underwent library construction and sequencing using the 10x Genomics platform. Downstream analysis involved quality control, normalization, dimensionality reduction (tSNE and UMAP), and cell type annotation using established marker genes from literature and databases. KEGG enrichment analysis was performed to compare the functional pathways identified by each method, with a specific focus on digestive enzyme-related genes.

Datasets

The study utilized pancreatic tissue samples from three healthy 10-day-old female Xiangdong black goats. The raw sequencing data generated from these samples are available in NCBI's Gene Expression Omnibus under the accession number GSE207644.

Key Insights and Ideas

  • scRNA-seq offers superior cell type resolution in the goat pancreas: The study found that scRNA-seq identified a greater number of distinct cell types, including pancreatic stellate cells, immune cells, and delta cells, which were not uniquely identified by snRNA-seq. Conversely, snRNA-seq uniquely identified pancreatic stem cells but also classified a significant portion of cells as "other" due to a lack of clear annotation.
  • Digestive enzyme gene expression is better captured by scRNA-seq: Genes related to digestive enzymes, crucial for pancreatic exocrine function, showed significantly higher expression levels when analyzed by scRNA-seq compared to snRNA-seq. This suggests scRNA-seq is more effective for profiling key functional genes in this tissue.
  • snRNA-seq's limitations in this context: The snRNA-seq data exhibited a higher proportion of reads mapping to the antisense strand and a lower sequencing saturation, potentially indicating a less complete representation of the transcriptome. This, coupled with the large "other" cell population, suggests snRNA-seq may be less suitable for comprehensive analysis of the goat pancreas under these experimental conditions.
  • Method choice is context-dependent: While scRNA-seq proved superior in this study, the authors emphasize that the optimal choice between scRNA-seq and snRNA-seq depends on factors like sample compatibility, technical considerations, and specific research objectives.

Key Results

  • Cell Counts: scRNA-seq identified 11,351 valid cells, while snRNA-seq identified 14,634 valid cells.
  • Gene Detection: scRNA-seq detected a median of 1615 genes per cell, while snRNA-seq detected a median of 1190 genes per cell.
  • Cell Type Annotation:
    • Both methods identified pancreatic acinar cells, ductal cells, alpha cells, beta cells, and endothelial cells.
  • scRNA-seq uniquely identified pancreatic stellate cells (63.09% of total), immune cells (2.81%), and delta cells (2.00%).
  • snRNA-seq uniquely identified pancreatic stem cells (1.55%) and had a large "other" cell population (56.18%).
  • Digestive Enzyme Gene Expression: Genes like CEL, CPB1, CTRC, PNLIPRP2, and PRSS2 showed consistently higher expression levels in scRNA-seq compared to snRNA-seq. For example, CEL expression was 135.36 in scRNA-seq versus 37.73 in snRNA-seq.

Limitations and Future Work

The authors acknowledge several limitations:

  • Limited Sample Size: The study used only three goats, which may limit the generalizability of the findings.
  • Technical Biases: Both scRNA-seq and snRNA-seq are susceptible to technical biases related to sample processing, cell capture efficiency, and RNA integrity. The reliance on amplification in scRNA-seq can introduce artifacts, and sample multiplexing might lead to batch effects.
  • Inter-sample Variability: The modest sample size also contributes to potential inter-sample variability.

Future work suggested includes:

  • Systematic Evaluation: Further studies should systematically compare both methods across diverse sample conditions (e.g., fresh vs. frozen tissues) to validate their complementary utilities.
  • Optimizing Protocols: Research should focus on optimizing tissue processing protocols and evaluating how pre-analytical variables like dissociation time and cold ischemia influence sequencing outcomes.
  • Integrated Application: The authors suggest that an integrated application of both scRNA-seq and snRNA-seq, rather than exclusive use of one, might optimize caprine pancreas studies.

Summary updated

Abstract

The objective of this study was to determine whether single-cell RNA sequencing (scRNA-seq) or single-nucleus RNA sequencing (snRNA-seq) was more effective for studying the goat pancreas. Pancreas tissues from three healthy 10-day-old female <i>Xiangdong black goats</i> were processed into single-cell and single-nucleus suspensions. These suspensions were then used to compare cellular composition and gene expression levels following library construction and sequencing. Both scRNA-seq and snRNA-seq were eligible for primary analysis but produced different cell identification profiles in pancreatic tissue. Both methods successfully annotated pancreatic acinar cells, ductal cells, alpha cells, beta cells, and endothelial cells. However, pancreatic stellate cells, immune cells, and delta cells were uniquely annotated by scRNA-seq, while pancreatic stem cells were uniquely identified by snRNA-seq. Furthermore, the genes related to digestive enzymes showed a higher expression in scRNA-seq than in snRNA-seq. In the present study, scRNA-seq detected a great diversity of pancreatic cell types and was more effective in profiling key genes than snRNA-seq, demonstrating that scRNA-seq was better suited for studying the goat pancreas. However, the choice between scRNA-seq and snRNA-seq should consider the sample compatibility, technical differences, and experimental objectives.
BibTeX
@article{cheng2025single,
  author = {Jie Cheng and Tianxi Zhang and Yan Cheng and Kefyalew Gebeyew and Zhiliang Tan and Zhixiong He},
  title = {{Single-Cell RNA Sequencing Outperforms Single-Nucleus RNA Sequencing in Analyzing Pancreatic Cell Diversity and Gene Expression in Goats}},
  journal = {International Journal of Molecular Sciences},
  year = {2025},
  doi = {10.3390/ijms26083916},
}
Figure