Analysis of single cell RNA sequencing (scRNA-seq) data

scRNA-seq analysis was performed as described previously [23].

In brief, mouse PDAC cells (1 × 10⁵ shNC or shLDHA cells) were suspended in 50 µL PBS, then orthotopically injected into the pancreas of 5-week-old female C57BL/6 mice. After 20 days, the tumors were sliced into small pieces (2–4 mm) on ice. Single cell suspensions were acquired using the Tumor Dissociation Kit (Miltenyi Biotec, 130–096-730) at 37 °C for 40 min. One sample was integrated from five orthotopic syngeneic PDAC tumors. CD45⁺ immune cells in the tumor tissues were sorted by FCM, then cell suspensions were centrifuged and resuspended in PBS containing 0.2% BSA for scRNA-seq.

The Chromium Next gel bead-in-emulsion (GEM) Single-cell 3' Reagent Kits v3.1 manual (10X Genomics, Pleasanton, CA, USA) was used to generate a library of 10⁴ cells/sample. The cDNA library was generated using a droplet-based sequencing platform (10X Genomics) and sequenced by DNBSEQ-G400 (MGI Tech, Shenzhen, China).

Raw sequencing data were aligned to the mm10 mouse reference genome and processed to a matrix containing normalized gene counts versus cells per sample by the Cell Ranger software package (version 5.0.0, 10X Genomics). The Seurat R package (version 4.3.0) was used to import and process the matrix. “CellCycleScoring”, “SCTransform”, and “Harmony (version 0.1.1)” were performed to eliminate any cell cycle influence or batch effects on the data [24]. The “DoubletFinder” R package (version 2.0.3) was used to filter the identified doublets [25]. The “RunPCA” function was used for principal component analysis and the “FindNeighbor” function was used to estimate clusters. Then, the “FindClusters” function by the Uniform Manifold Approximation and Projection (UMAP) method was used for visualization.