Human CD34+ cell isolation
Human fetal liver (FL) samples were obtained from Advanced Biosciences Resources (Alameda, CA) with proper consent. FL samples were cut in small fragments, treated for 25 min at 37 °C with Collagenase D (100 ng/mL; Roche). The cell suspension was prepared, and the human CD34+ cells were separated by density gradient centrifugation, followed by positive immunomagnetic selection using anti-human CD34 microbeads according to the manufacturer’s instructions (Miltenyi Biotec). Cells were either frozen in 10% DMSO containing human albumin serum and stored in liquid nitrogen or injected directly.
Mice
Immunodeficient mice with Rag2 and IL-2Rγ deletions were generated and bred in-house. Two lines were used: SRG: Balb/cAnN;129S4/SvJae – Rag2–/− human Sirpα IL-2Rγ−/− and SRG-15: Balb/cAnN;129S4/SvJae – Rag2−/−, human Sirpα, IL-2Rγ−/−, human IL-15. Mice were maintained on a sulfa diet (LabDiet, St. Louis, MO) in a MPF facility, and were intra-bred for about 10-11 generations. All mice in the study were 5–7 months of age at the start of the experiment. A mix of male and female were used for all experiments. All experiments were done in compliance with Regeneron Institutional Animal Care and Use Committee protocols.
Human CD34+ immune cell reconstitution
The generation of knock-in mice encoding human SIRPa And IL-15 in a 129xBALC/c genetic background were created by Velocigene technology. Mice were crossed to a Rag2−/− IL2Rg−/− background and will be abbreviated as follow: SRG (=Sh/mRG) and SRG-15 (= Sh/mRG-15h/m). Newborn pups were sublethally irradiated (360 cGy; X-RAD 320 irradiator) 4–24 h prior to an intrahepatic injection of 1 × 105 human FL-derived CD34+ cells.
Analysis of human hematopoietic cell populations
Blood was collected retro-orbitally 10–12 weeks after engraftment. Red blood cells were lysed using ACK (Gibco) and the cells were stained with the following monoclonal antibodies from Invitrogen and BD for flow cytometry analysis. For overall hematopoietic engraftment: anti-mouse CD45-APC-Cy7 (30-F11, 1:100), anti-human CD45-PE-Cy5.5 (HI30, 1:100), anti-human CD19-FITC (HIB19, 1:100), anti-human CD3-PB (S4.1, 1:20), anti-human NKp46-APC (9E2, 1:20), anti-human CD14-PE-Cy7 (M5E2, 1:33) and anti-human PD-1-BV605 (EH12, 1:40). The samples were acquired by a Fortessa (BD Biosciences) or Symphony (BD Biosciences) and analyzed using FACSDiva (BD Biosciences) and FlowJo software. Mice with ≥10% hCD45+ of total circulating CD45+ cells (total including both mouse and human CD45+ cells) were used for experiments. For experimental repeats, different donor sources of human CD34+ cells were used. Donor-to-donor variations were comparable with the range of variation between individual same donor CD34+ cell-engrafted mice.
Cell culture
Raji (CCL-86; ATCC) and Ramos (CRL-1596; ATCC) were grown as monolayer cultures in RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum (Gibco), penicillin (100 units/mL), streptomycin (100ug/mL), and l-glutamine (200 mM) at 37 °C in a humidified, 10% CO2 incubator. Raji and Ramos tested negative by PCR for mouse viruses and mycoplasma. Cells were authenticated by STR profiling and were 100% matched to the database.
Tumorigenesis
HIS-reconstituted and non-HIS-reconstituted SRG-15 mice were randomly assigned to treatment groups. Overall, 2 × 105 tumor cells were implanted subcutaneously under anesthesia in the right flank of the mouse. Volume was determined by calipher measurement and calculated using the following formula: tumor volume (mm3) = 0.5 × (length × width2). Tumor measurements were collected twice a week and evaluated as progressor (tumor growth only), regressor (two tumor volume values smaller than the preceding peak value) and rejection (two tumor volume values where the second value is zero). Progressor group: P; Combined regressor and rejection groups: R.
T-cell depletion
Mice were injected intra-peritoneally with 200 μg of anti-CD4 (OKT4; BioXcell) and/or 200 μg anti-CD8 (OKT8; BioXcell) monoclonal antibodies at days −2, −1, 0, +1, +6, and +14 relative to Raji tumor implantation. Control mice were given 200 μg mouse IgG2b and/or mouse IgG2a isotype control at same time as anti-CD4 and/or anti-CD8 treatment. Tumor growth was measured then as described above.
ELISPOT assay
ELISPOT assay was established to determine the frequency of T cells capable of responding to a specific stimulus by secretion of IFNγ. Each well in a 96-well PDVF ELISPOT plate (Millipore) was coated with hIFNγ capture antibody(BD) at 1:200 of phosphate-buffered saline (PBS), pH 7.2 (Gibco). Following overnight incubation at 4 °C, the plate was blocked with 10% fetal calf serum (Gibco) in PBS for 2 h at 37 °C. Splenic T cells were isolated by immunomagnetic selection using anti-human microbeads according to the manufacturer’s instructions (Easy Sep Human T Cell Isolation Kit; StemCell Technologies). Purified T cells were added 2 × 105 per well in triplicates. Tumor cells were added to selected wells at 4:1 T-cell/tumor cell ratio. In some experiments, tumor cells were pre-blocked with 25 μg/mL of pan-MHC class I (clone W6/32; Biolegend) and/or pan-MHC class II (clone Tu83; Biolegend) blocking antibodies. Plate was incubated overnight at 37 °C in a humidified, 10% CO2 incubator. Plate was extensively washed in 0.05% Tween 20–PBS and hIFNγ detection (BD) was added to each well at 1:250 of 10%FBS + PBS for two hours at room temperature. Following an extensive washed with 0.05% Tween 20–PBS the wells were coated with Streptavidin-HRP (MABTECH) at 1 μL per mL of 10% FBS + PBS for 45 min at room temperature. The plate was washed with 0.05% Tween 20–PBS extensively. TMB substrate (MABTECH) was added to each well for 10–20 min, until the color developed. Substrate reaction was stopped by washing wells with water. Plate was dried, and the spots were enumerated by AID plate reader and the data was analyzed by GraphPad Prism.
Tissue homogenization
Raji tumors were minced, and enzymatically digested for 20 min with intermittent agitation using tumor dissociation kit (Miltenyi, 130-096-730) at 37 °C, treated with RPMI and filtered (70-mm nylon filter; Falcon). Spleen was harvested and homogenized. Single-cell suspensions of tumor and spleen were subjected to ACK lysing buffer (Gibco). Tumor and spleen single-cell suspensions were incubated with anti-mouse CD45-APC-Cy7 (30-F11, 1:100), anti-human CD45-PE-Cy5.5 (HI30, 1:100), anti-human CD19-FITC(HIB19, 1:100), and anti-human CD3-PB (S4.1, 1:20). The cells were sorted on Astrios (Beckman Coulter) for human T cells. Sorted cells were used for single-cell sequencing.
Single-cell preparation and sequencing
Single cells suspended in PBS with 0.04% BSA (~6000 cells) were loaded on a Chromium Single Cell Instrument (10X Genomics). RNA-seq and V(D)J libraries were prepared using Chromium Single Cell v1.0 5’ Library, Gel Beads & Multiplex Kit (10X Genomics). After amplification, cDNA was split into RNA-seq and V(D)J library aliquots. To enrich the V(D)J library aliquot for TCR a/b, the cDNA was split into two 20 ng aliquots and amplified in two rounds using primers designed in-house. Specifically, for first round amplification the primers used were MP147 (ACACTCTTTCCCTACACGACGC) for short R1, MP122 (GGTGCTGTCCTGAGACCGAG) for mouse TRAC, and MP123 (CAATCTCTGCTTTTGATGGCTCAAAC) for mouse TRBC. For second round amplification, 20 ng aliquots from the first round were amplified using MP147 (ACACTCTTTCCCTACACGACGC) for short R1, MP130 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTTGGTACACAGCAGGTTCTGG) a nested R2 plus mouse TRAC, and MP131 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTGACCTTGGGTGGAGTCACATTTCTC) a nested R2 plus mouse TRABC. In the case of human TRAC and TRBC, the procedure was the same as above, but used MP120 (GCAGACAGACTTGTCACTGGA) and MP121 (CTCTGCTTCTGATGGCTCAAACA) for round 1 enrichment and MP128 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT GCAGGGTCAGGGTTCTGGATA) and MP129 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT ATG GCT CAA ACA CAG CGA CCT) for round 2 enrichment. V(D)J libraries were prepared from 25 ng each mTRAC or hTRAC and mTRBC or hTRBC amplified cDNA. Paired-end sequencing was performed on Illumina NextSeq500 for RNA-seq libraries (Read 1 26-bp for unique molecular identifier (UMI) and cell barcode, 8-bp i7 sample index, 0-bp i5, and Read 2 55-bp transcript read) and V(D)J libraries (Read 1 150-bp, 8-bp i7 sample index, 0-bp i5, and Read 2 150-bp read). For RNA-seq libraries, Cell Ranger Single-Cell Software Suite (10X Genomics, v2.2.0) was used to perform sample demultiplexing, alignment, filtering, and UMI counting. The mouse mm10 genome assembly and RefSeq gene model for mice were used for the alignment. For V(D)J libraries, Cell Ranger Single-Cell Software Suite (10X Genomics, v2.2.0) was used to perform sample demultiplexing, de novo assembly of read pairs into contigs, align and annotate contigs against all of the germline segment V(D)J reference sequences from IMGT, label and locate CDR3 regions, group clonotypes.
Singe-cell data analyses
Single-cell RNA-seq data QC Single cells were filtered for downstream analysis by the following criteria: UMI (unique molecular identifier) count within the range between 500 and 6000 and mitochondria percentage greater than 20% of the total UMI count. 271 cells were filtered out. Gene expression (in UMI) is scale normalized to log1p(UMI.Count*scaling.factor/(total UMI count)), where scaling.factor = 10,000 and log1p is log(1 + x). The variable genes were found with the average expression between 0.0125 and 3.0, with a dispersion greater than 0.5.
All experiments were duplicated. We used the dataset from one experiment for discovery and the other for result validation. For principal component analysis (PCA), the analysis was run on normalized and transformed UMI counts on variably expressed genes. These PC outputs were loaded as input to generate t-SNE plots. Fourteen clusters were defined based on the selected clustering resolution. 3D t-SNE plot was drawn using Rtsne package with default parameters. 2D t-SNE plot was generated with Seurat package. Differentially expressed genes were identified for clustering based on Avg. |log(fold change)| > 0.25 (or, avg. fold change > 1.28 or <0.78) and Wilcoxon rank-sum test (adj. P value <0.01) to only include genes that are detected in at least 10% of cells in either of the two populations.
TCR cloning
Tumor-infiltrating lymphocytes and human T cells were isolated from the tumor and matching spleen, respectively. The splenic T cells and TILs were sorted for CD3 + TCR + cells. Raji-specific TCRs in expanded T cells were identified using TCR repertoire sequencing data. TCR expression constructs were generated by synthesizing gBlock gene fragments (IDT) encoding bicistronic cassettes with the full-length TCRA sequence followed by TCRB, separated by a F2A self-cleaving peptide sequence (full insert sequences are provided in Supplementary Table 1). Fragments were cloned into the pLVX-EF1a-IRES-Zeocin lentiviral vector (Takara) via XbaI and NotI restriction sites.
TCR reporter lines
TCR sequences were introduced by lentiviral transduction into a parental Jurkat reporter line lacking endogenous TCRα/β expression (via Crispr-mediated disruption), and engineered to express human CD8α, CD8β, and an AP1-responsive luciferase reporter (Qiagen # CLS-011L). Lentivirus was generated by co-transfection of TCR, Gag/Pol, and VSV-G envelope expression constructs into HEK-293T cells (ATCC CRL-3216) using Lipofectamine LTX reagent (Life Technologies 15338100). Cell culture supernatants were harvested 48 h post-transfection, and lentivirus was concentrated using the Lenti-X reagent (Takara 631232) according to the manufacturer’s protocol. Transduction of Jurkat cells was done by adding 20 μl concentrated viral supernatant to 2 × 105 cells in 200 μl total media volume in round-bottom 96-well plates, and spinning plates at 2500 rpm for 90 min at RT. Transduced cells were enriched by antibiotic selection. Surface TCR+ CD8+ CD28+ cells were FACS-sorted and expanded under antibiotic selection media. The reactivity of the sorted clones reactivity was measured against a dose titration of Raji, Ramos, HEK293, K562-HLA-A3, K562-HLA-B15 and K562-HLA-CW3 (# of APCs to 50,000 Jurkat reporter cells) by using One-Glo luciferase assay kit (Promega) according to manufacturer’s instructions.
Statistics and reproducibility
Tumor implantation experiments were repeated in at least ten different HSC donors. A minimum of 5–7 mice were used in each experimental group. ELISpots had triplicate wells per condition and student t tests (unpaired) were performed to compare conditions with a P value of less than 0.05, indicating statistical significance.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
