Antigen-induced chimeric antigen receptor multimerization amplifies on-tumor cytotoxicity

Abstract

Ligand-induced receptor dimerization or oligomerization is a widespread mechanism for ensuring communication specificity, safeguarding receptor activation, and facilitating amplification of signal transduction across the cellular membrane. However, cell-surface antigen-induced multimerization (dubbed AIM herein) has not yet been consciously leveraged in chimeric antigen receptor (CAR) engineering for enriching T cell-based therapies. We co-developed ciltacabtagene autoleucel (cilta-cel), whose CAR incorporates two B-cell maturation antigen (BCMA)-targeted nanobodies in tandem, for treating multiple myeloma. Here we elucidated a structural and functional model in which BCMA-induced cilta-cel CAR multimerization amplifies myeloma-targeted T cell-mediated cytotoxicity. Crystallographic analysis of BCMA-nanobody complexes revealed atomic details of antigen-antibody hetero-multimerization whilst analytical ultracentrifugation and small-angle X-ray scattering characterized interdependent BCMA apposition and CAR juxtaposition in solution. BCMA-induced nanobody CAR multimerization enhanced cytotoxicity, alongside elevated immune synapse formation and cytotoxicity-mediating cytokine release, towards myeloma-derived cells. Our results provide a framework for contemplating the AIM approach in designing next-generation CARs.

Fig. 1

Synergistic BCMA interactions with the cilta-cel tandem of nanobodies. a Schematic diagrams of BCMA and CAR constructs used in this study. ECD extracellular domain, TM transmembrane region, ICD intracellular domain, LTR long terminal repeat, SP signal peptide, Nb1, Nb2, and NbT, nanobody 1, 2, and tandem. Regarding the CAR constructs, the SP, hinge, and TM domains are derived from human CD8α. b, c Characteristics of BCMA–nanobody binding assayed by surface plasmon resonance (SPR). SPR sensorgrams of nanobodies as the analytes (concentration [nM] indicated) over immobilized BCMA_ECD-His (b) or BCMA_ECD-Fc (c) as the ligands. Binding models (Langmuir or heterogeneous ligand) used for fitting (black overlay) are denoted above and the derived affinities and kinetics are shown below each chart

Fig. 2

Crystallographic analysis of the BCMA–nanobody complexes. a Surface and cartoon representation of the crystal structure of Nb1 in complex with BCMA_ECD. Nb1 is colored blue and BCMA green. The putative N-linked glycosylation sites of BCMA (N42) are highlighted in pink. Inset, electron density (2F_O − F_C map contoured at 1 σ) of the amino (N)-terminal region of BCMA is shown as gray mesh. b Cartoon representation of the Nb1 complex with the complementarity-determining region (CDR) 1 of Nb1 colored in green, CDR2 in cyan, and CDR3 in orange while framework region (FR) 1 in purple blue, FR2 in pink, FR3 in yellow, and FR4 in marine. c Open-book views of BCMA complexes showing interfaces with antibodies and ligands. Interface residues are outlined in red or orange. d, e Surface and cartoon representation of the crystal structure of Nb2 in complex with BCMA_ECD. Nb2 chains are colored wheat and light wheat while BCMA molecules pale green and praseodymium green for the observed two pairs in the asymmetric unit (d). The putative N-linked glycosylation sites of BCMA are highlighted. For the cartoon representation in (e), the Nb2 CDRs and FRs are color-coded as in (b). f An open-book view showing the interface between Nb2 and BCMA_ECD. The interface residues are outlined in red

Fig. 3

Interdependent BCMA–nanobody multimerization in solution. a Analysis of the molecular masses of the BCMA_ECD, nanobodies, and NbT in solution using sedimentation velocity analytical ultracentrifugation (SV-AUC). The derived molecular weights are indicated for the corresponding peaks. Cartoon illustrations reflect our interpretation of the stoichiometries. b Molecular masses of the BCMA_ECD complexes assayed by SV-AUC. Excess BCMA_ECD was used to form the nanobody complexes

Fig. 4

A model for the BCMA–NbT complex. a Superposition of the Nb1 and the Nb2 complexes based on BCMA_ECD. The secondary interface observed in the Nb1–BCMA_ECD structure is omitted for clarity. b Nanobody epitopes mapped on the BCMA_ECD surface. Residues in contact with Nb1 are outlined in blue while those in contact with Nb2 in yellow. c Competition for BCMA binding between Nb1 and Nb2 assayed using SPR. Fc-tagged human BCMA_ECD was immobilized onto a Protein A sensor chip as the ligand. Nb1 (2000 nM) or Nb2 (100 nM) was loaded with BCMA_ECD-Fc for 1 min, followed by a mixture of Nb1 (2000 nM) and Nb2 (100 nM). d Small-angle X-ray scattering (SAXS) envelopes and corresponding models of Nb1 (left), Nb2 (middle), and NbT (right) based on the crystal structures of the Nb1–BCMA_ECD and the Nb2–BCMA_ECD complexes

Fig. 5

Soluble nanobodies inhibit ligand binding and signaling of BCMA. a Superposition of the APRIL (PDB: 1XU2) and BAFF (PDB: 1OQD) complexes with our nanobody complexes based on BCMA. Neighboring APRIL chains that sandwich BCMA are colored pink and light pink, while BAFF chains are in aquamarine and pale cyan. b APRIL and BAFF interface compared to Nb1 and Nb2 epitopes on BCMA. Color coding is explained above the BCMA models. c Nanobodies inhibit APRIL-induced MM.1S proliferation. Data are represented as means and standard errors of the mean (SEM) of n = 3 replicates per group, the one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test was used to assess the differences among different groups. Data are represented as mean ± SEM. d Nb1 (2000nM; left), Nb2 (100 nM; middle), and NbT (10 nM; right) were probed for competitive binding against APRIL using SPR

Fig. 6

Antigen-induced CAR dimerization enhances on-tumor cytotoxicity. a Viability of MM.1S-luc (upper) and RPMI 8226-luc cells (lower) co-cultured with T cells transduced with the indicated nanobody CAR constructs compared to untreated (UT) control T cells (color coding at the top of the panel) for 24 h at the effector cell: target cell (E: T) ratios of 1:1 and 4:1. n = 4 replicates per group. Unpaired two-tailed Student’s t-test was performed on each grouped sample without adjustments for multiple comparisons. b Granzyme A, granzyme B, and perforin levels in the supernatant were measured. n = 4 replicates per group. Unpaired two-tailed Student’s t-test was performed on each grouped sample without adjustments for multiple comparisons. c Representative imaging flow cytometry micrographs of tested CAR T–MM.1S interactions at ×40 magnification. d CAR T cells were assessed for mean fluorescence intensity of F-actin at the immune synapse. n = 4 replicates per group. Unpaired two-tailed Student’s t-test was performed without adjustments for multiple comparisons. e Evaluation of cell-surface BCMA levels on CHO-luc cells. Escalating amounts of BCMA RNA were delivered by electroporation and the antibody binding capacity was assessed using an anti-BCMA antibody conjugated with phycoerythrin. f Viability of CHO-luc cells electroporated with varying BCMA amounts as measured in (e) while being co-cultured with the indicated CAR T cells for 24 h at the effector cell: target cell (E: T) ratios of 4:1. The one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test or the Kruskal-Wallis test followed by Dunn’s multiple comparisons correction was performed to assess the differences among different groups, n = 3 replicates per group. All data are represented as mean ± SEM

Fig. 7

Model of BCMA-induced nanobody CAR multimerization