Dnase 1

Prior to euthanasia, mice were anaesthetized with 6.9 μl/g of a Zoletil-mix (2.4 mg/ml Zolazepam, 2.4 mg/ml Tiletamin, 3.8 mg/ml Xylazin, 0.095 mg/ml Buturphanol). 3 min before euthanasia, 250 μl of anti-CD45.2 – fluorescein isothiocyanate (BD Pharmingen, clone 104, 1:100 dil.) were intravenously injected into the tail of the mice to label vascular leukocytes. For euthanasia, mice were exposed to CO₂ (3 L/min for 5–10 min).
Samples were obtained from 4 to 12 mice per group (individually or pooled in groups of 2) in RPMI 1640 (Gibco Invitrogen). Single-cell suspensions were created by homogenizing organs through a 100 μm nylon filter (Falcon). In addition. GTs were incubated before homogenization for 45 min. at 37 °C CO₂ with type IV collagenase (0.8 mg/ml) (Sigma) and DNAse I (Roche) (0.08 mg/ml). Before and after incubation GTs were processed with gentleMACS™ Dissociator (Miltenyi Biotec) before mechanical filtering. Cell suspensions were centrifuged (700 × g, 5 min) and washed twice in RPMI 1640. Cell pellets from all organs were resuspended in RPMI-1640 (Gibco Invitrogen) supplemented with 5 ×10⁻⁵ M 2-mercaptoethanol, 1 mM glutamine, 1% pyruvate, 1% penicillin-streptomycin, 1% HEPES, and 10% FCS (Gibco Invitrogen). An additional filtration step was performed after resuspension of GT samples to further eliminate adhesive cellular debris and mucus.

Three samples, each of tumor and adjacent normal tissue, from GC patients were collected. Clinical details of these patients are provided in Table S2. Samples 4, 5, and 6 were used for scRNA-seq extraction and sequencing. The tissue samples were cut into small pieces and then digested in RPMI-1640 at 37 °C for 30 min using 1 mg/mL of Type IV collagenase (17104019, Thermo Fisher, USA) and 50 µg/mL of DNase I (10104159001, Roche, Germany). Following digestion, the tissue underwent filtration through a 70 µm cell strainer to obtain a single-cell suspension, followed by treatment with trypsin. Single cells were captured using the C1 Single-Cell Auto Prep System (Fluidigm, Inc., South San Francisco, USA). Once captured, the cells were lysed within the chip to release mRNA, followed by reverse transcription to generate cDNA. The lysed and reverse-transcribed cDNA underwent pre-amplification within a microfluidic chip for subsequent sequencing. The amplified cDNA was used for library construction and subjected to scRNA-seq on the HiSeq 4000 Illumina platform (parameters: paired-end reads, read lengths of 2 × 75 bp, ~20,000 reads per cell)^{63 (link)}.
Data analysis was conducted using the “Seurat” package in R software. Data were quality-controlled based on the criteria of 200 < nFeature_RNA < 5000 and percentage of mitochondrial genes <20, with the top 2000 highly variable genes selected for further analysis^{64 (link)}.
To reduce the dimensionality of the scRNA-seq dataset, we performed principal component analysis (PCA) based on the top 2000 highly variable genes. The top 20 principal components (PCs) were chosen for downstream analysis using the Elbowplot function in Seurat. Cell subpopulations were identified using the FindClusters function with a default resolution setting res=1. Subsequently, nonlinear dimensionality reduction was performed on the scRNA-seq data using the t-SNE algorithm. Markers for various cell subpopulations were selected using the Seurat package, and cell annotations were performed using the “SingleR” package^{65 (link)}.
Cell-cell communication analysis was carried out using the “CellChat” package in R. Differentially expressed genes (DEGs) within the scRNA-seq dataset were identified using the “Limma” package in R. DEGs between normal and GC samples were filtered based on the criteria |logFC| > 0.5 and adjusted P-value < 0.05^{66 (link)}.