Natural high-avidity T-cell receptor efficiently mediates regression of cancer/testis antigen 83 positive common solid cancers

Background T-cell receptor-engineered T cells (TCR-Ts) have achieved encouraging success in anticancer clinical trials. The antigenic targets, however, were primarily focused on human leukocyte antigen (HLA) A*02:01 restricted epitopes from a few cancer/testis antigens (CTAs) which are not widely expressed in common solid cancers; the tested T-cell receptors (TCRs) were frequently from tumor-infiltrating lymphocytes of old patients and were not assured to have higher avidity. Here, we propose the isolation of high-avidity TCRs against CTAs that are frequently expressed in common solid cancers. Methods We selected the CT83 protein, which is frequently expressed in common solid cancers, as a model antigen for screening of its specific TCR. The predicted CT83 epitopes with strong or weak binding to HLA-I molecules, popular in the Chinese population, were integrated into three synthetic long peptides. CT83 reactive CD8+ T cells were stimulated with peptide-loaded dendritic cells (DCs) and sorted using the CD137 biomarker for single-cell sequencing to obtain the paired TCRαβ sequence. The higher frequency TCRs were reconstructed for characterization of the CT83 epitope and for assessment of in vitro and in vivo antitumor activities. Results CT83 reactive T cells from young healthy donors (YHDs) were generated by repeated stimulation with DCs and peptides. The single-cell TCR sequencing results of reactive T cells indicated that a single TCR clonotype dominated the paired TCRs. T cells engineered with this dominant TCR led to HLA-A*11:01-restricted recognition of the CT8314-22 epitope, with higher avidity. Functional assays showed powerful cytotoxicity in vitro against the targets of several CT83-positive solid cancer cell lines. Furthermore, TCR-Ts showed therapeutic efficacy in three xenograft solid tumor models. The meta-analysis of gene expression of 92 CTAs indicated that most CTAs did not or at low levels in the thymus, which suggested that those CTAs may experience incomplete thymic central tolerance. Conclusions High-avidity TCR against CT83 could be isolated from YHDs and efficiently mediate regression of well-established xenograft common solid tumors. The high-avidity TCR repertoire in the peripheral blood of some donors for CT83 and other CTAs provides the basis for the efficient isolation of high-avidity TCRs to target numerous solid cancers.

was conducted in the BD FACSAria II. The data were processed using FlowJo software and the gating strategy is shown in Supplemental Figure 5.
Single cell TCR sequencing.
CD3 + CD8 + CD137 + T cells (2 × 10 5 ) were sorted from above peptide-reactive T cells and subjected to single cell RNA sequencing (scRNA-seq) (10x Genomics, USA). The sequencing data were processed for identification of TCR clonotype by alignment and annotation using the Cell Ranger pipeline (version 2.1.0).

CRISPR/Cas9 gene editing.
To disrupt the endogenous HLA-I expression of TAP-deficient T2 cells (HLA-A2+), we conducted CRISPR/Cas9 B2M gene editing experiment to target "GAGTAGCGCGAGCACAGCTA" DNA sequence of human B2M gene. Briefly, 6.25 μg Cas9 protein (Invitrogen, A36498) and 37.5 pmol sgRNA were mixed to form a B2M ribonucleoprotein complex (B2M-RNP) and added into 100 μL Nucleofector buffer (Lonza). 1×10 6 T2 cells were transfected with the above B2M-RNP/ Nucleofector solution by electroporator 2B Nucleofector with A-030 program. HLA-I expression was detected by staining with anti-HLA-ABC antibodies (Biolegend, 311406). Pure HLA-I negative cells were enriched by sorting with FACS AriaII (BD) and limiting dilution method. To establish the CT83 negative tumor cell line, we constructed a KO plasmid (the structure diagram is shown Supplemental Figure1A containing CT83 sgRNAs and Cas9 coding sequence, the sgRNA target sequences are respectively "TGGCTAAAATATACTTACTG"and"AGGCGGTACTAAGTGCCGCC". 1 × 10 6 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) cells of SGC7901 were transfected with the CT83 KO plasmid by electroporator 2B Nucleofector. The inclusion of the EGFP and Puro genes in the KO plasmid allowed to monitor the transfection under a fluorescence microscope and enrich the CT83 negative cells (SGC7901 CT83 KO ) by treating with puromycin.

Western blot analysis.
CT83 protein expression of tumor cell lines was analyzed by standard western blotting with cell lysate prepared by Tissue and Cell Total Protein Extraction Kit (Applygen, Beijing, China). The blotting membrane was detected using anti-huCT83 mouse monoclonal antibody (CL4762, AB219971, Abcam) (1:1000) and goat-anti-mouse IgG conjugated with horseradish peroxidase (HRP) (BS12478, Bioworld) (1:40000); GAPDH was used as normalized control. The protein bands were detected by ECL Plus Western Blotting Substrate (Boster Bio, CA, USA) and recorded on an X-ray film, the density of bands was determined with Image J software.

IFN-γ ELISPOT assay.
The number of peptide-reactive T cells was determined using ELISPOT assay. Briefly, analyzed by Immuno Spot v6.0 software with the recommended parameters.

CTA bioinformatics analysis.
To identify the optimal target antigens of CTAs, we obtained 224 CTA genes from GenBank and searched the GTEx Portal database (https://www.gtexportal.org/home/ index.html) to analyze their expressions with strict criteria for specific expression in testis: 1. Strictly no expression in vital organs/tissues including brain and heart: protein-transcript per million (pTPM) ≤0.1; 2. Slight expression in not more than one non-vital tissue type except testis: pTPM >1.0 ~ < 3.0. Next, we searched TCGA

Cross-reactive epitope analysis.
After inferring the essential non-anchor residues in the verified epitope, the potential cross-reactive peptides of TCR1 were analyzed by the online tool Expitope2.0 (http://webclu.bio.wzw.tum.de/expitope2/). The peptides in human proteins were analyzed for the mismatches compared to the verified epitope and predicted for the binding affinity with HLA-A*11:01. The CT83 KO plasmid: two CT83 sgRNAs, Cas9 coding sequence, the EGFP and Puro genes are included. (B) CT83/TCR1 retroviral vector. The paired TCR1 ab V genes were respectively linked with mouse TCR ab C genes and cloned into pMP71 retroviral vector. A P2A cleavage sequence was added between the TCR chains, and a T2A-linked delta-LNGFR (CD271) gene was included as a biomarker for monitoring of TCR expression.  Figure 7. Study of the off-target effect of TCR induced by homologous epitope By Expitope2.0 search for candidate peptides that shared at least 4/9 residues were tested for CT83/TCR1-T recognition BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)