Fig. 2: Yeast surface display identifies CD19 CAR binding mimotopes.
a, Illustration of CD19 CAR stimulation by natural CD19 or amph-mimotope. Created in BioRender. Ma, L. (2025) https://BioRender.com/miyfij7. b, Yeast surface display workflow for identifying antibody-specific mimotopes. Created in BioRender. Ma, L. (2025) https://BioRender.com/5ugsnwl. c, Flow cytometry plots showing the successful identification of yeast cell populations (P1 and P2) binding to 1 µM FMC63IgG after a single round of magnetic enrichment. d, Structural modelling of mimotopes P1 and P2 binding to FMC63. e, Effect of the disulfide bond on mimotope binding to FMC63IgG on the yeast surface. Yeast cells bearing mimotope P1 were treated with increasing concentrations of DTT and then stained with 500 nM biotin-FMC63IgG and an anti-HA tag antibody for analysis of binding by flow cytometry. f, ELISA analysis showing the impact of the disulfide bond on synthetic mimotope binding to FMC63IgG. Created in BioRender. Ma, L. (2025) https://BioRender.com/xd8q66t. g, Scheme for amph-mimotope generation by click chemistry linkage of azide-modified P1 to DBCO-PEG-DSPE, and schematic illustration of amph-P1 coating of target cells. Created in BioRender. Ma, L. (2025) https://BioRender.com/5ugsnwl. h, CD19− K562 cells labelled with 100 nM amph-P1, unlabelled K562 cells or CD19+ NALM6 cells were co-cultured with CD19 CAR-T at a 5:1 E:T ratio for 6 h followed by IFNγ ELISA (****P < 0.0001). i, K562 cells unlabelled, labelled with 100 nM amph-P1 or labelled with 100 nM amph-pepvIII were co-cultured with control EGFRvIII-CAR-T cells at a 5:1 E:T ratio for 6 h followed by IFNγ ELISA. Error bars show mean ± s.d. with triplicate samples. ****P < 0.0001 by one-way ANOVA with Tukey’s post-test.
