Single-Cell RNA Sequencing of Lung Tissue in a Rat Model of Acute Respiratory Distress Syndrome

Summary

This article presents a standardized protocol for single-cell RNA sequencing (scRNA-seq) analysis of lung tissue in a rat model of Acute Respiratory Distress Syndrome (ARDS). It details steps from LPS-induced model establishment to bioinformatic analysis, identifying 21 cell clusters and 6 macrophage subpopulations, and highlights key inflammatory pathways involved in ARDS pathogenesis.

Abstract

Acute Respiratory Distress Syndrome (ARDS) is a severe inflammatory lung condition with high mortality, characterized by dysregulated immune responses and alveolar damage, yet its cellular heterogeneity remains incompletely understood. Single-cell RNA sequencing (scRNA-seq) offers unprecedented resolution to dissect cell-type-specific mechanisms, but its application in ARDS is hampered by technical variability. Here, we present a standardized protocol for scRNA-seq analysis of lung tissue in a lipopolysaccharide (LPS)-induced rat ARDS model. This method includes optimized tissue dissociation, rigorous quality control, doublet removal, batch effect correction, and comprehensive bioinformatic pipelines for clustering, annotation, and differential expression analysis. Our approach reliably identifies 21 distinct cell populations and reveals six macrophage/monocyte subpopulations with unique transcriptional signatures in ARDS. Subcluster-specific differential gene expression and functional enrichment analysis further identified key genes and pathways underlying ARDS pathogenesis. This reproducible workflow enables in-depth characterization of cellular diversity, inflammatory pathways, and candidate therapeutic targets, providing a robust foundation for mechanistic and translational studies in ARDS.

Introduction

Acute Respiratory Distress Syndrome (ARDS) is an acute respiratory illness characterized by diffuse lung inflammation and non-cardiogenic pulmonary edema, which primarily manifests as progressive hypoxemia and respiratory failure^1,2. The pathological mechanisms of ARDS involve dysregulated inflammatory cascades, alveolar-capillary barrier disruption, and abnormal activation and interactions of multiple immune cell types^3,4. During ARDS progression, immune cells within lung tissue drive inflammatory cascades through secretion of inflammatory mediators and modulation of immune responses, thereby disrupting pulmonary endothelium, triggering activation of the extrinsic coagulation cascade, and exacerbating lung injury^5,6,7. However, current understanding of cellular-level pathological mechanisms in ARDS remains limited, which restricts the development of specific pharmacologic therapies, while current management strategies primarily consist of various modalities of oxygen support and mechanical ventilation¹. In-depth characterization of cellular heterogeneity and dynamic changes during ARDS pathogenesis is of paramount importance for identifying novel therapeutic targets and developing new treatment strategies^8,9,10.

The emergence of single-cell RNA sequencing (scRNA-seq) technology has brought revolutionary breakthroughs to RNA sequencing research, enabling us to dissect cellular heterogeneity and functional dynamics in disease pathogenesisn^11,12. Compared to traditional bulk RNA sequencing, scRNA-seq technology can identify rare but critical cell subpopulations, reveal dynamic processes of cell-state transitions, and discover complex intercellular interaction networks, information that is often masked in population-level analyses^13,14,15. Recent scRNA-seq studies in ARDS have achieved significant advances, including the identification of specific monocyte gene expression signatures, the discovery of abnormal macrophage polarization patterns, and the elucidation of profibrotic intercellular communication networks^16,17,18,19. These findings not only deepen our understanding of the pathophysiological mechanisms of ARDS but also provide new insights into biomarker development, disease stratification, and precision therapy^20,21.

However, current ARDS scRNA-seq research faces challenges, including insufficient methodological standardization, substantial differences in sample processing protocols, and inconsistent data analysis pipelines, which limit the reproducibility of research results and their clinical translational applications^22,23. This protocol addresses the technical challenges associated with scRNA-seq analysis of ARDS lung tissue by optimizing procedures for tissue processing, quality control, and data analysis, thereby providing a foundation for mechanistic studies and therapeutic development. The standardized approach ensures reproducibility across different laboratories and facilitates comparative studies in ARDS research of cellular heterogeneity.

Protocol

Twelve specific pathogen-free (SPF) male Sprague-Dawley (SD) rats, aged 6–7 weeks and weighing 220 g ± 10 g, were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Animal License Number: SCXK (Jing) 2025-0008). The rats were housed in the Laboratory Animal Center of Dongzhimen Hospital, Beijing University of Chinese Medicine, in compliance with the national standard “Laboratory Animal-Requirements of Environment and Housing Facilities” (GB 14925-2010, issued by the National Laboratory Animal Standardization Technical Committee of China). The experimental procedures were approved by the Animal Welfare and Ethics Committee of Dongzhimen Hospital, Beijing University of Chinese Medicine (Approval Number: 24-54). The reagents and the equipment used are listed in the Table of Materials.

1. Model establishment and sample collection

NOTE: Prior to the experiment, subject all rats to a 1-week acclimatization period to minimize stress-induced effects.

1. Dissolve Escherichia coli lipopolysaccharide (LPS) in 0.9% saline to obtain a 0.4 mg/mL LPS solution.
2. Randomly divide 12 SPF male Sprague-Dawley (SD) rats into two groups: a control group (n = 6) and an ARDS model group (n = 6).
3. Administer 0.9% saline via injection intratracheally at a dose of 0.5 mL/100g body weight to the control group.
4. Administer the 0.4 mg/mL LPS solution via injection intratracheally at the same volume (0.5 mL/100g) to the ARDS model group, and after 16 h, anesthetize the rats using isoflurane (following institutionally approved protocols).
5. Collect blood via the abdominal aorta, then centrifuge (4 °C, 1000 × g, 30 min) to obtain serum.
6. Open the thoracic cavity and extract lung tissue.
7. Perform bronchoalveolar lavage via the bronchial stump using a syringe to obtain Bronchoalveolar Lavage Fluid (BALF)^24,25.
8. Harvest the right lung lobe and rinse it with PBS for subsequent pathological examination.
9. Harvest the left lung lobe and rinse it with PBS for subsequent single-cell sequencing and molecular biology experiments^26,27.

2. Histopathological assessment of lung tissue

1. Immerse the harvested right lung tissue of rats in 4% paraformaldehyde for fixation for 12 h.
2. Following fixation, dehydrate the tissue through a graded series of ethanol (70%, 80%, 90%, 95%, 100%, and 100%), allowing 2 h at each concentration.
3. Treat the dehydrated tissue with xylene until it becomes transparent.
4. Infiltrate the transparent tissue with molten paraffin in a 60 °C oven for 2 h.
5. Place the infiltrated tissue block into an embedding mold, pour in fresh paraffin, and allow it to cool and solidify into a paraffin block.
6. Section the paraffin block into 4-μm thick slices using a microtome, float the sections on a 45 °C water bath to spread them, and then pick them up with poly-L-lysine-coated slides.
7. Bake the slides in a 60 °C oven to ensure firm adhesion of the sections to the slides.
8. Deparaffinize the sections by immersion in xylene and rehydrate them through a descending gradient of ethanol (100%, 95%, 90%, 80%, and 70%).
9. Rinse the sections with PBS, perform hematoxylin and eosin (HE) staining, and mount the coverslips using neutral resin.
10. Assess the lung injury on whole-slide scanned sections using a standardized scoring system based on neutrophil infiltration, interstitial inflammation, edema, and congestion.
    NOTE: Each parameter was scored from 0 to 4 based on the extent of pathological involvement: 0 (none); 1 (≤25%); 2 (26-50%); 3 (51-75%); 4 (>75%)^28,29.

3. Enzyme-linked immunosorbent assay

NOTE: Samples may be maintained at -80 °C if analysis cannot be performed immediately after acquisition.

1. Dissolve the standards and allow them to stand at room temperature for 15 min, then perform serial dilution according to the manufacturer's instructions.
2. Prepare the biotinylated antibody working solution, enzyme conjugate working solution, and wash buffer as instructed in the manual.
3. Add 100 μL of each sample or 100 μL of serially diluted standards to the antibody-coated wells, and incubate at 37 °C for 90 min.
4. Wash the plate 4 times using a plate washer, add 100 μL of biotinylated antibody working solution, and incubate at 37 °C for 30 min.
5. Wash the plate 4 times, add 100 μL of enzyme conjugate working solution, and incubate at 37 °C for 30 min.
6. Add 100 μL of substrate solution per well, incubate protected from light at 37 °C for 10 min.
7. Add 100 μL of stop solution per well, mix gently, and measure the optical density (OD) at 450 nm using a microplate reader.
8. Generate a standard curve by plotting the OD values against standard concentrations using the designed software.
9. Determine sample concentrations based on the measured OD values and the standard curve.

4. Preparation of single-cell suspensions for sequencing

NOTE: After lung tissue is obtained, it should be placed on ice and processed promptly to maintain cell viability.

1. Wash the lung tissue 3 times with PBS pre-cooled to 4 °C.
2. Add 2 mL of pre-cooled PBS and further mince them into smaller fragments using sterile scissors into approximately 0.5 mm³ in size.
3. Use a Pasteur pipette to transfer the minced tissue together with PBS into a 15 mL centrifuge tube.
4. Add an additional 1 mL of PBS pre-cooled to 4 °C to the culture dish, rinse and resuspend the remaining tissue fragments, and transfer the suspension to the same tube.
5. Adjust the volume to 6.5 mL with PBS pre-cooled to 4 °C, digest tissue fragments in enzyme solution containing Collagenase I (100 U/mL) and Dispase II (1 U/mL), and incubate with gentle shaking at room temperature for 30 min.
6. When the tissue fragments become translucent, and no distinct solid pieces remain, filter the digested tissue through 70 μm and 40 μm cell strainers, then centrifuge (4° C, 500 × g, 5 min) and discard the supernatant.
7. Immediately resuspend the pellet in 3 mL of red blood cell (RBC) lysis buffer, mix gently by pipetting, and lyse on ice for 10 min.
8. When the supernatant appears pale yellow or nearly colorless, add 7 mL of PBS to terminate the reaction, centrifuge (4 °C, 500 × g, 5 min), and discard the supernatant.
9. Add 3 mL of PBS to resuspend the cells and determine cell concentration and viability using trypan blue staining.
   NOTE: If cell viability is below 70%, perform a dead cell removal procedure.
10. Wash the cells twice with a washing buffer containing 0.04% bovine serum albumin (BSA) prepared in DPBS.
11. Resuspend the cells for counting and adjust the concentration to 1,000 cells/μL for subsequent loading and detection.

5. Quality control for single-cell sequencing data

NOTE: Single-cell RNA libraries were prepared from cell suspensions using a droplet-based single-cell library construction method. Libraries were sequenced on a high-throughput platform using DNA nanoball–based sequencing chemistry. Approximately 21,000 cells were loaded per library, aiming for a cell recovery rate of ~60%. Sequencing generated a total of 5.0 × 10⁸ reads per library, with an average depth of 30,000-40,000 reads per cell. Raw sequencing data were processed using the corresponding single-cell RNA analysis pipeline to produce the gene expression matrix for each sample.

1. Construct the Oligo library by PCR and barcode labeling, ensuring concentration >10 ng/μL and peak size 180 bp ±10 bp.
2. Construct the cDNA library by fragmentation, end repair, adaptor ligation, and PCR, ensuring concentration >10 ng/μL and peak size 350–550 bp.
3. Generate a Seurat object for each sample from the gene expression matrix using the Read10X function in the Seurat package (v4.3.0) in R software (v4.2.0).
4. Remove low-quality cells by filtering out those with mitochondrial gene proportion >10% or hemoglobin gene proportion >5%.
5. Normalize the data using the "LogNormalize" method with a scale factor of 10,000.
6. Identify highly variable features using the "vst" method and retain the top 2,000 genes for downstream analysis.
7. Perform cell cycle scoring using Seurat's built-in cell cycle marker genes and regress out cell cycle effects (G2M.Score and S.Score) during data scaling.
8. Detect and remove potential doublets using DoubletFinder (v2.0.3) with pN = 0.25, pK = 0.09, and an expected doublet rate of 0.06, and exclude these from further processing.
9. Correct batch effects across samples using the Harmony package (v0.1.1) with theta = 2 and max.iter.harmony = 20.
10. Add information columns for individual sample names and grouping details of the ARDS model group and control group to the metadata in the Seurat object.

6. Dimensionality reduction, clustering, and cell type annotation

1. Perform principal component analysis (PCA) on highly variable genes using the RunPCA function in Seurat.
2. Determine the optimal number of principal components through ElbowPlot visualization.
3. Identify major cell clusters by applying the Leiden algorithm via Seurat's FindClusters function and FindNeighbors function with a resolution of 0.1.
4. Conduct UMAP dimensionality reduction using the selected principal components.
5. Annotate cell types by identifying marker genes with the FindAllMarkers function.
6. Visualize marker gene expression across cell types using the DotPlot function.
7. Add a new information column of cell annotation results to the metadata and visualize cell type annotation with Umap.
8. Calculate the proportion of each experimental group within each identified cell type using dplyr functions to count cells and compute percentages.
9. Generate stacked bar plots with ggplot2 to illustrate group composition per cell type, including percentage labels and statistical comparisons between groups.

7. Subpopulation analysis

1. Extract the target cell subpopulation using the subset function and create a corresponding Seurat object for downstream subpopulation-level analysis.
2. Re-normalize and re-scale the subpopulation data while regressing out mitochondrial genes, hemoglobin genes, UMI counts, and cell cycle scores.
3. Perform secondary PCA analysis and apply Harmony batch correction specifically for the subpopulation.
4. Conduct re-clustering using FindNeighbors and FindClusters functions to identify distinct subtypes within target cell subpopulations.
5. Generate new UMAP embeddings and visualize subpopulation clusters with cell type annotations following the same approach as previously described.

8. Differential gene expression analysis

1. Identify differentially expressed genes (DEGs) between the ARDS model group and control group for each subcluster using the MAST statistical method.
2. Incorporate latent variable regression to control for sequencing depth effects during differential expression testing.
3. Filter DEGs based on pre-defined log fold change thresholds and adjusted p-values using the FindMarkers function.
4. Generate volcano plots with the ggplot function for each subcluster to visualize differential gene expression patterns with significance labeling.
5. Export DEG results and create comprehensive summary tables to facilitate functional enrichment and other downstream analyses.

9. Enrichment analysis

1. Map gene symbols to ENTREZ IDs using the org.Rn.eg.db annotation database for pathway analysis compatibility.
2. Combine DEGs from all subclusters and filter significant genes based on log fold change thresholds and adjusted p-values (Wilcoxon rank-sum test, logfc.threshold = 0.25, p_val < 0.05).
3. Perform Gene Ontology (GO) enrichment analysis using enrich GO function with Biological Process, Molecular Function, Cellular Component ontology categories, and Benjamini-Hochberg correction method.
4. Conduct Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis using the enrich KEGG function with the rat organism database and and apply statistical significance thresholds.
5. Generate dot plots for GO and KEGG enrichment results using the dotplot function, displaying the top 15 significantly enriched terms with gene ratios and adjusted p-values.
6. Export enrichment results as CSV files and save visualization plots for comprehensive pathway analysis interpretation.

Results

HE staining of rat lung tissue showed clear alveolar structures with minimal inflammation in controls, whereas the ARDS model group exhibited distorted alveolar cavities, thickened walls, and significant inflammatory cell infiltration, as shown in corresponding lung injury scores (Figure 1A–E). Elevated levels of IL-6, TNF-α, and IL-1β in BALF and serum confirmed successful modeling and a systemic inflammatory response (Figure 2A–F

Discussion

This protocol presents a standardized approach for single-cell RNA sequencing analysis of lung tissue in a rat model of acute respiratory distress syndrome (ARDS). These results demonstrate the successful establishment of the LPS-induced ARDS model, as confirmed by histopathological examination, elevated lung injury scores, and increased levels of inflammatory cytokines in BALF and serum. Single-cell transcriptomic analysis revealed 21 distinct cell populations, with notable increases in macrophages/monocytes, neutrophil...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.4% Trypan blue solution	Gibco, USA	15250061
4% paraformaldehyde	Servicebio, China	G1101
Bovine serum albumin (BSA)	Sangon Biotech, China	A600332-0005
Collagenase I	YEASEN, China	40507ES60
CurveExpert	Hyams Development	v1.4
Dispase II	YEASEN, China	40104ES60
Escherichia coli lipopolysaccharide (LPS)	Sigma, USA	L6511
hematoxylin and eosin (HE) staining	Solarbio, China	G1120
High-throughput gene sequencer	MGI Tech Co., Ltd	MGISEQ-2000RS
Phosphatase inhibitor	Beyotime Biotechnology, China	P1081
Phosphate buffered solution	Solarbio, China	P1020
PMSF	Solarbio, China	R0010
R software	Lucent Technologies, USA	v4.2.0
rat IL-1β ELISA kit	4abio, China	CRE0006
rat IL-6 ELISA kit	4abio, China	CRE0005
rat TNF-α ELISA kit	4abio, China	CRE0003
Red blood cell (RBC) lysis buffer	Solarbio, China	R1010
Red blood cell lysis buffer	Tiangen Biotech, China	RT122-1
RNase-free water	Ambion, USA	AM9937
TE buffer, pH 8.0	Ambion, USA	AM9858
Tris-Glycine Transfer Buffer	Servicebio, China	G2017