Droplet-based scRNA-seq analysis

Droplet-based scRNA-seq analysis of lymph nodes stromal cells

Reagents for Isolation

• PBS

• RPMI-1640 medium (Sigma-Aldrich, cat. no. R8758)

• Fetal Calf Serum (Biowest, cat. no. S1810-050)

• Penicillin-streptomycin (Lonza, cat. no. DE17-602E)

• HEPES buffer solution (PAN Biotech, cat. no. P05-01500)

• EDTA-2Na (Sigma-Aldrich, cat. no. E5134)

• Collagenase P (Roche, cat. no. 12138730001)

• DNase I (Sigma-Aldrich, cat. no. 4527)

• Dispase (Roche)

Equipment

• Flow cytometer

• Constant-temperature incubator/shaker

• 15-ml tube (BD Falcon, cat. no. 352096)

• 50-ml tube (BD Falcon, cat. no. 352070)

• 1.5 ml tube (Biomedical sciences, cat. no. BC-MPF-150C)

• 5-ml polystyrene round-bottomed FACS tubes (BD Biosciences, cat. no. 352058)

• 1 ml cut tips

• 24 well culture plates (TPP, cat no. 009224)

• 3 ml syringes (BD Plastipak)

• 26 G needles (Terumo)

• 70 μm nylon cell strainer (BD Falcon, cat. no. 340336)

Prepare:

1. Digestion medium

   RPMI 2% of FCS

   20mM HEPES

2. Digestion medium + enzymes

   Dispase (40 µg/ml)

   Collagenase P (1 mg/ml)

   DNAse (25 µg/ml)

   24 well plate with 1ml of Digestion medium without enzymes/well. One well is needed per lymph node.

    

For droplet based RNAseq

Chromium Single Cell 3’ Reagent Kit (v2 Chemistry)

10X Chromium Controller (10X Genomics)

Preparation of stromal cells

1. Preparation of lymph node stromal cells

2. Sacrifice the mice and harvest the lymph nodes. Transfer them in to the 24 well plate containing digestion medium without enzymes.

3. Replace the medium without enzymes with 1 ml of with pre-warmed digestion medium + enzymes (37°C).

4. Disrupt the lymph nodes using 26 G needles to make small pieces.

5. Transfer the disrupted organs in to a fresh 15ml Falcon tube using cut tips.

6. Incubate 15 min in 37°C water bath; do not incubate cells for longer.

7. During the incubation, prepare a 2nd round of fresh 15mL Falcon tubes. Add 5mL of MACS buffer (2% FCS, 3mL of EDTA in PBS 1X) and put the tubes on ice.

8. After incubation, disrupt gently the tissue by pipetting up and down using 1 ml cut tips. Wait 2-3 minutes to let the nondigested tissue go to the bottom. Transfer the supernatant to the tube containing MACS buffer. Refill the tubes with 1mL of pre-warmed digestion medium + enzymes.

9. Incubate 10 min in the water bath and repeat step 7 until no pieces of tissue are visible. Normally it takes three rounds of digestion to get a single cell suspension. After the second round of digestion, normal 1 ml tips can be used for tissue dissociation.

10. Centrifuge the tubes 1200 RPM 5 min.

11. Discard the supernatant and resuspend the pellet in 1ml of MACS buffer.

12. Add 50 µL of anti-CD45 and anti-Ter119 MACS beads.

13. Incubate 20 min on ice.

14. Wash with 5 ml of MACS buffer and centrifuge at 1200 RPM 5 min.

15. In the meantime, activate the MACS columns with 3 ml of MACS buffer. Add collection tubes in the columns. Important the negative population contains the stromal cells.

16. Resuspend the cells in 5mL of MACS buffer and add the cells to the top of the MACS column.

17. Wash 2 times with 3 ml of MACS buffer. Collect the negative population (Stroma cells)

18. Centrifuge the negatively selected cells (CD45 ter119 depleted cells) 1200 RPM 5 min. Discard the supernatant.

19. Wash the cells by adding 5 ml of MACS buffer. Centrifuge the samples at 1200 RPM for 5 min. discard the supernatant.

20. Cells are ready for staining and cell sorting.

21. For the study, single cell suspension of stromal cells were stained with anti-Ter119, anti-CD19 and anti-CD45 antibodies and the viability dye 7AAD. EYFP⁺ Ter119^-, CD45^-, CD19^-, 7AAD^- were sorted.

22. After sorting, cells were processed on a 10X Chromium Controller (10X Genomics) following the recommended protocol for the Chromium Single Cell 3’ Reagent Kit (v2 Chemistry). Generated cDNA libraries were sequenced on a NovaSeq 6000 sequencing system from Illumina at the Functional Genomic Center in Zurich.

23. First processing of sequencing files was performed using CellRanger (v3.0.1) with Ensembl GRCm38.94 release as reference followed by an additional quality control using scater R/Bioconductor package (v1.11.2) running in R v3.6.0. Damaged cells were removed based on UMI counts, total counts of detected genes (more than 2.5 median absolute deviations from the median across all cells) and mitochondrial gene content (more 2.5 median absolute deviations above the median across all cells). Finally, cells expressing one of the genes Lyve1, Prox1 or Pecam1 were removed as contaminating cells and only cells expressing mRNA encoding for EYFP were kept for downstream analysis resulting in a total of 15,436 cells and 23,983 genes expressed in more than one cell.

    Following quality control data was analyzed using functions from the Seurat package (v3.1.1) for normalization, scaling, dimensional reduction with PCA and UMAP and graph-based clustering. Resulting clusters were characterized and compared across conditions based on marker genes and differentially expressed genes inferred from Wilcoxon test as implemented in the FindMarkers function (Seurat v3.1.1). In order to further summarize functional differences between conditions all genes were ranked based on a signal-to-noise ratio statistic that was calculated on normalized expression values. Ranked gene lists were tested for enrichment of gene sets from the mSigDB (v7.0) C2 collection with GSEA-Preranked (v7.0.3) in a GenePattern notebook. Immune signatures were derived from the top differentially expressed genes that were grouped according to immune-related functions. In order to analyze and compare the immune activation process in FRCs across different conditions cells were ordered along a diffusion map that was computed based on all genes included in one of the immune signatures by running the runDiffusionMap function from the scater R/Bioconductor package (v1.14.1)

Category