Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jan 10;120(2):e2216352120.
doi: 10.1073/pnas.2216352120. Epub 2023 Jan 4.

Inefficient exploitation of accessory receptors reduces the sensitivity of chimeric antigen receptors

Jake Burton  1 , Jesús A Siller-Farfán  1 , Johannes Pettmann  1 , Benjamin Salzer  1 , Mikhail Kutuzov  1 , P Anton van der Merwe  1 , Omer Dushek  1
Affiliations

Inefficient exploitation of accessory receptors reduces the sensitivity of chimeric antigen receptors

Jake Burton et al. Proc Natl Acad Sci U S A. .

Abstract

Chimeric antigen receptors (CARs) can redirect T cells to target abnormal cells, but their activity is limited by a profound defect in antigen sensitivity, the source of which remains unclear. Here, we show that CARs have a > 100-fold lower antigen sensitivity compared to the T cell receptor (TCR) when antigen is presented on antigen-presenting cells (APCs) but nearly identical sensitivity when antigen is presented as purified protein. We next systematically measured the impact of engaging important T cell accessory receptors (CD2, LFA-1, CD28, CD27, and 4-1BB) on antigen sensitivity by adding their purified ligands. Unexpectedly, we found that engaging CD2 or LFA-1 improved the antigen sensitivity of the TCR by 125- and 22-fold, respectively, but improved CAR sensitivity by only < 5-fold. This differential effect of CD2 and LFA-1 engagement on the TCR vs. CAR was confirmed using APCs. We found that sensitivity to antigen can be partially restored by fusing the CAR variable domains to the TCR CD3ε subunit (also known as a TRuC) and fully restored by exchanging the TCRαβ variable domains for those of the CAR (also known as STAR or HIT). Importantly, these improvements in TRuC and STAR/HIT sensitivity can be predicted by their enhanced ability to exploit CD2 and LFA-1. These findings demonstrate that the CAR sensitivity defect is a result of their inefficient exploitation of accessory receptors and suggest approaches to increase sensitivity.

Keywords: T cells; adhesion receptors; antigen sensitivity; chimeric antigen receptors.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interest.

Figures

Fig. 1.
Fig. 1.
CARs show reduced sensitivity compared to the TCR when antigen is presented on APCs but not when presented as purified protein. (A) Schematic of antigen receptors. The 1G4 TCR and the D52N scFv both recognize the 9V NY-ESO-1 peptide antigen presented on HLA-A*02:01. CARs using the CD8a hinge contain the CD8a transmembrane domain, whereas CARs using the IgG1 or CD28 hinges contain the CD28 transmembrane domain. (B) Schematic of APC stimulation system. (C and D) Representative dose–response showing (C) cytotoxicity by LDH release and (D) surface expression of CD69 for the TCR and the indicated CARs along with EC50 values from at least three independent experiments determined by fitting a Hill function to each dose–response curve. (E) Representative dose–response when the purified biotinylated 9V pMHC ligand is presented on streptavidin-coated plates (Left two plots) and EC50 values from at least three independent experiments (Right). The EC50 values are compared using (C and D) one-way ANOVA or (E) one-sample t-test for a hypothetical mean of 1.0 on log-transformed values. *P-value ≤ 0.05, **P-value ≤ 0.01, ***P-value ≤ 0.001, ****P-value ≤ 0.0001.
Fig. 2.
Fig. 2.
Systematic engagement of accessory receptors identifies that CARs are inefficient at exploiting the adhesion receptors CD2 and LFA-1 relative to the TCR. (A) Schematic of accessory receptors and their ligands. (B and C) Representative dose–response curves showing T cell activation by upregulation of surface CD69 measured by flow cytometry after 24 h using the solid-phase stimulation assay. T cells were presented with purified pMHC alone (“None”) or with a fixed concentration of 250 ng/well of the indicated accessory receptor ligand (colors) for the (B) TCR and (C) the indicated CARs. (D) The EC50values for the indicated antigen receptor and purified ligand condition were obtained by fitting a Hill function to each dose–response curve. Individual EC50values for each antigen receptor are from an independent experiment (N ≥ 3). The numbers indicate the fold-change in EC50induced by the accessory receptor ligand relative to pMHC alone (“None”), and statistical significance is determined by a paired t-test on log-transformed data. (E) The data in (D) are presented in a different format showing the fold-change in EC50between the TCR and the indicated CAR for pMHC alone or the indicated accessory receptor ligand. The fold-change is compared using a one-sample t-test to a hypothetical value of 0 on log-transformed data. *P-value ≤ 0.05, **P-value ≤ 0.01, ***P-value ≤ 0.001, ****P-value ≤ 0.0001.
Fig. 3.
Fig. 3.
Abrogating the CD2 and LFA-1 adhesion interaction disproportionately impacts the antigen sensitivity of the TCR compared to the CAR. (A) Schematic of CD58 and ICAM-1 blocking experiment on the HLA-A2+ glioblastoma U87 target cell line. (B) Representative dose–response curves for the indicated blocking conditions for the TCR (Left) and CAR (Right). (C and D) Fold-change in EC50between the CAR and TCR (Left) or relative to the isotype (Right) for (C) CD69 and (D) 4-1BB upregulation. (E) Schematic of CD58 and ICAM-1 knockout experiments. (F) Representative dose–response curves for the indicated target cell lines for the TCR (Left) and CAR (Right). (G and H) Fold-change in EC50between the CAR and TCR ((Left) or relative to the isotype (Right) for (G) CD69 and (H) 4-1BB. Individual EC50values for CD69 or 4-1BB are determined by a fit to the dose–response curve from at least three independent experiments (each data point in C, D, G, andH is from an independent experiment). The fold-change between the TCR and CAR is compared using a two-sample t-test to the isotype or parental line condition (Left in C, D, G, and H) or directly between the TCR and CAR (Right in C, D, G, and H) on log-transformed values. *P-value ≤ 0.05, **P-value ≤ 0.01, ***P-value ≤ 0.001, ****P-value ≤ 0.0001.
Fig. 4.
Fig. 4.
The ability of TCR-like chimeric antigen receptors to recapitulate the sensitivity of the TCR depends on the efficiency with which they are able to exploit the CD2 adhesion interaction. (A) Schematic of “TCR-like” engineered antigen receptors. (B–E) T cells expressing the indicated antigen receptor were cocultured with T2 target cells pulsed with different peptide antigen concentrations for 8 h. Representative dose–response (Top) and fitted EC50values from at least three independent experiments (Bottom) are shown for (B and C) cytotoxicity (measured by LDH release) and (D and E) CD69 upregulation. (F–I) T cells expressing the indicated antigen receptor were stimulated by a titration of purified pMHC alone (solid lines) or in combination with a fixed concentration of purified CD58 (dashed lines). Representative (F and H) dose–response curves and (G and I) fitted EC50values from at least three independent experiments for (F and G) CD69 upregulation and (H and I) IFNγ production. (J) The averaged EC50values for CD69 upregulation from the APC stimulation assay (from panel C) are plotted over the averaged fold-change in EC50for CD69 induced by the addition of CD58 from the solid-phase stimulation assay (from panel G). The EC50values are compared using a one-way ANOVA on log-transformed values (C, E, G, and I). *P-value ≤ 0.05, **P-value ≤ 0.01, ***P-value ≤ 0.001, ****P-value ≤ 0.0001.
Fig. 5.
Fig. 5.
Adhesion receptors more efficiently enhance antigen engagement for the TCR compared with the eTruC and CAR. (A and B) Effect of pMHC with or without CD58 or ICAM-1 on surface antigen receptor expression, as determined by pMHC tetramers. (A) Representative curves and (B) fitted IC50 values from at least three independent experiments. All data are normalized to surface expression without pMHC. All comparisons are made using a one-way ANOVA on log-transformed values. *P-value ≤ 0.05, **P-value ≤ 0.01, ***P-value ≤ 0.001, ****P-value ≤ 0.0001. (C) Model showing similar performance of a TCR and a CAR in the absence of ligands for adhesion receptors (Left) and superior enhancement of TCR vs. CAR antigen engagement (red arrows) and signaling (blue arrows) by adhesion receptor ligands (Right).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. June C. H., O’Connor R. S., Kawalekar O. U., Ghassemi S., Milone M. C., CAR T cell immunotherapy for human cancer. Science 359, 1361–1365 (2018). - PubMed
    1. Exley A. R., McBlane J., Regulating innovation in the early development of cell therapies. Immunother. Adv. 1, 1–18 (2021).
    1. Fry T. J., et al. , CD22-targeted CAR T cells induce remission in B-ALL that is Naive or resistant to CD19-targeted car immunotherapy. Nat. Med. 24, 20–28 (2018). - PMC - PubMed
    1. Park J. H., et al. , Long-term follow-up of CD19 CAR therapy in acute lymphoblastic leukemia. N. Engl. J. Med. 378, 449–459 (2018). - PMC - PubMed
    1. Shah N. N., Fry T. J., Mechanisms of resistance to CAR T cell therapy. Nat. Rev. Clin. Oncol. 16, 372–385 (2019). - PMC - PubMed

Publication types

MeSH terms

Substances