Cell hashing antibodies

Jurkat cells were “stained” with decreasing concentrations (1:100, 1:500, 1:1000) of Cell Hashing antibodies (BioLegend, USA; B2M, CD298 pool) as described above and passed through a 10x Genomics Single Cell workflow to yield ~ 2000 cells. As a control, non-hashed cells were passed through a separate 10x Genomics Single Cell lane. Cells from both experiments were subsampled to the same numbers of cells and reads per cell to compare UMI and gene counts.

Cells were stained with Aqua live/dead stain (Thermo Fisher, Inc., Waltham, MA, USA), and live cells were sorted using FACS Aria or the Sony MA900 Multi-Application cell sorter to separate live cells with robust (CFSE^low), moderate (CFSE^med), and no proliferative response (CFSE^high) to the TCR stimulus (Figure 1A). CFSE^low, CFSE^med, and CFSE^high cells were incubated with cell hashing antibodies (Biolegend, Inc., San Diego, CA, USA) TotalSeq™-B0252, TotalSeq™-B0253, and TotalSeq™-B0254, respectively (please refer to Supplementary Methods for details). After staining, cells were mixed back together, 10,000 from each population, and 12,000 total cells were loaded into the Chromium Controller, aiming to achieve the targeted recovery of 10,000 cells in the scRNA-Seq experiment. Reverse transcription to generate cDNA and library preparation for scRNA-Seq were conducted per the manufacturer’s instructions (10X Genomics, Inc., Pleasanton, CA, USA) using v3 kits. Sequencing was conducted using the NovaSeq 6000 instrument (Illumina, Inc., San Diego, CA, USA) at the Institute of Genomics Medicine (IGM) Genomics Center, University of California San Diego. The data were provided in the form of paired .fastq files.

CITE-seq data was generated from 30 samples (10 individuals, 3 experimental conditions) by multiplexing to reduce batch effects and costs using cell hashing (49 (link)) (for experimental conditions) and genotype-based demultiplexing (27 (link)) (for individual donors). For CITE-seq experiments, antibody-oligo conjugates against 39 surface proteins (e.g., CD8, CD45RA, CD4, CD3, CCR7, CD69) as well as the Cell Hashing Antibodies (HTOs) were purchased from BioLegend. PBMCs from different donors were independently stained with one of our HTO-conjugated antibody pools (based on the experimental condition) and a pool of 39 immunophenotypic markers for CITE-seq. Equal numbers of cells from all 30 samples were pooled and super-loaded onto 10 lanes of 10× Genomics Single Cell 3P v2 (10× Genomics, USA), yielding approximately 25 K cells per lane. All libraries were sequenced together on Illumina NovaSeq S4 flowcells.