Directing CAR NK Cells via the Metabolic Incorporation of CAR Ligands into Malignant Cell Glycans

Abstract

The abundance of sialic acid-containing glycans in the glycocalyx of malignant cells enables immune evasion. Here, we leverage the biosynthetic pathways that permit pervasive sialylation to incorporate a chimeric antigen receptor (CAR) ligand into malignant cell glycans, and demonstrate that this increases the susceptibility of malignant cells to the cytolytic activity of CAR-expressing natural killer (NK) cells. Specifically, we applied a C-9-functionalized nonnatural sialic acid [i.e., fluorescein sialic acid (FL-SA)] to modify malignant cell glycans. We confirm the metabolic incorporation of FL-SA into plasma membrane-associated glycans. The preparation of anti-fluorescein CAR NK cells permitted studies demonstrating that treating malignant cells with FL-SA increased susceptibility to CAR NK cell-mediated cytolysis. Furthermore, we observed that the specificity of the anti-fluorescein CAR NK cells is enhanced for fluorescein-labeled cells, and an increased release of cytokines from the CAR NK cells upon incubation with FL-SA-treated cells. The results arising from this study demonstrate that CAR ligands can be metabolically incorporated into malignant cells, and we reason that such strategies could be leveraged to tackle the issue of antigen heterogeneity that limits the clinical efficacy of CAR T/NK cell therapies.

Figure 1.

NK cell interactions with malignant cells. (a) Sialoglycan-mediated immune evasion—the binding of Siglecs-7 and −9 on NK cells to sialoglycans on a malignant cell leads to inhibitory signaling that dampens NK cell activation. (b) Sensitizing malignant cells to CAR NK cell-mediated cytolysis—the metabolic labeling of a malignant cell with fluorescein via the sialoglycan biosynthetic pathway enables the recognition of the malignant cells by an anti-fluorescein CAR NK cell, CAR NK cell activation, and the cytolysis of the malignant cell.

Figure 2.

FL-SA incorporation and Siglec binding. (a) Structure of FL-SA (1). (b) Increase in the cell surface levels of fluorescein on the surface of Raji cells after treatment with FL-SA (0.03–1 mM) for 48 h was determined by flow cytometry using an anti-fluorescein antibody. The fold increase in median fluorescence intensity relative to the vehicle control is reported. Significance is determined relative to vehicle control. (c) Persistence of fluorescein on the surface of Raji cells after treatment with FL-SA (1 mM) for 48 h was determined by flow cytometry. (d) Cell surface levels of fluorescein on Raji cells after treatment with FL-SA (1 mM) +/− sialyltransferase inhibitor 3Fax-Neu5Ac(OAc)₄ for 48 h. Significance is determined relative to the vehicle control. (e) Metabolic incorporation of FL-SA in a panel of cell lines. Significance is determined relative to the vehicle control. (f,g) Binding for the Siglec-7/9 Fc chimera to Raji cells treated +/− FL-SA (1 mM) for 48 h. Statistical significance was determined as described in the Supporting Information.

Figure 3.

CAR design and CAR NK cell preparation. (a) Designed construct containing the promoter, CAR, P2A peptide, and the selection markers is depicted. (b,c) Cell surface expression of the CAR was determined by flow cytometry using an APC-labeled anti-Myc tag antibody (clone 9E10). Statistical significance was determined as described in the Supporting Information.

Figure 4.

Interactions between FL-SA-treated Raji cells and CAR NK cells. (a) Cytolysis of Raji cells treated +/− FL-SA (1 mM) by CAR NK or WT NK cells at a range of effector to target ratios (E/T). Significance is determined relative to the WT + control Raji group. (b) Formation of conjugates between CAR NK cells and Raji cells treated +/− FL-SA (1 mM) was determined by flow cytometry (solid bars). In addition, the ability of a fluorescein analogue (2) to competitively inhibit conjugate formation was assessed (dashed bars). (c) Cytolysis of Raji cells treated +/− FL-SA (1 mM) by CAR NK or WT NK cells (E/T = 4:1) over time. Significance is determined relative to the WT + control Raji group. (d) Secretion of cytokines upon incubation of CAR NK or WT NK cells with Raji cells treated +/− FL-SA (1 mM) was determined using a cytometric bead array and is reported relative to NK cells incubated in the absence of target cells.

Figure 5.

Cytolysis of FL-SA-treated cell lines. (a–k) Cytolysis of target cells treated +/− FL-SA (1 mM) by CAR NK or WT NK cells at a range of effector to target ratios (E/T). Significance is determined relative to the WT + control target group. (l) CAR-specific lysis was calculated by subtracting the percent cytolysis in the WT + control target group from the percent cytolysis in the CAR + FL-SA target group at the highest E/T applied. Statistical significance is determined as described in the Supporting Information.