Fig 2: Structure, binding sites, and mode of action of half-life extended CD38-specific nanobody-based bispecific killer cell engagers (HLE-nano-BIKEs). (A) Scheme of bispecific and trispecific killer cell engagers (BiKEs, TriKEs) based on scFvs (top) or nanobodies (bottom). Tumor-specific and CD16-specific modules are indicated in yellow/orange and grey, respectively. Half-life extension (HLE) can be mediated by fusion to an Fc-fragment (top) or an albumin-specific nanobody (white, bottom). The paratope of each variable domain (corresponding to the CDR loops) is shown in red. Nano-TriKEs could be generated in the future by fusion of a nano-BiKE to a second tumor-specific nanobody or IL-15. (B) Scheme of CD38-specific HLE-nano-BiKEs (45 kDa) consisting of three nanobodies linked via flexible glycine-serine linkers. The N-terminal nanobody in each CD38-specific HLE-nano-BiKE recognizes one of three distinct epitopes of CD38: WF211 (pink, epitope 1 [E1]), MU1067 (blue, epitope 2 [E2]), and JK36 (green, epitope 3 [E3]). The central nanobody (grey) recognizes CD16 (FcγIII receptor on NK cells) and the C-terminal nanobody recognizes albumin (white). Conventional human antibody daratumumab (150 kDa) is indicated in yellow. (C) Scheme of the binding sites of daratumumab and the three HLE-nano-BiKEs. WF211-based HLE-nano-BiKE E1 recognizes an epitope (E1) that overlaps with that of daratumumab, MU1067-based HLE-nano-BiKE E2 and JK36-based HLE-nano-BiKE E3 bind independent epitopes (E2, E3). (D) Scheme of the proposed mode of action of a CD38-specific HLE-nano-BIKE. The N-terminal nanobody (WF211, MU1067, or JK36) binds CD38 on the myeloma cell, the central nanobody binds and activates an NK cell by targeting CD16, and the C-terminal nanobody extends the half-life of the construct by binding to albumin.

Fig 3: CD38-specific HLE-nano-BiKEs specifically induce NK cell-mediated cytotoxicity toward CD38+ myeloma cells. (A) Effector NK92 cells were pre-labeled with eFluor450 (eF450). Gating was performed to exclude cellular debris (lower left corner) of panel 1, and then on target cells, identified as eF450-negative cells (panel 2). Dead cells were excluded by PI-staining. (B) CD38KO GFP+ LP-1 myeloma cells labeled with eFluor670 (eF670) (upper population) were mixed with unlabeled CD38-expressing GFP+ LP-1 cells (lower population) and the indicated HLE-nano-BiKEs before addition of eF450-labeled NK92 cells stably transfected with human CD16 (lower panels) or lacking CD16 (upper panels, negative control) at an effector-to-target ratio of 3:1. Cells were incubated for 3 h at 37°C and analyzed by flow cytometry in the presence of the DNA staining dye propidium iodide (PI). Loss of cells from the alive cells was used to monitor HLE-nano-BiKE-induced lysis of target cells. Gating was performed on target cells as described in (A) Numbers indicate percentage of cells in each quadrant. Results are representative of three similar experiments.

Fig 4: Purified CD38-specific HLE-nano-BiKEs bind specifically and simultaneously to myeloma cells, NK cells, and albumin. (A) Purified HLE-nano-BiKEs and daratumumab (1 µg per lane) were size fractionated by SDS-PAGE under reducing conditions and visualized by Coomassie staining. (B) LP-1 luc myeloma cells (left panel) and NK92 hCD16 cells (right panel) were incubated with CD38-specific HLE-nano-BiKE E1, HLE-nano-BiKE E2, HLE-nano-BiKE E3, or an isotype control (HLE-nano-BiKE co). Bound HLE-nano-BiKEs were detected by flow cytometry using biotinylated human albumin and PE-Cy7-conjugated streptavidin by flow cytometry. Control stainings (unstained, dashed line) were performed with albumin and PE-Cy7-conjugated streptavidin alone. Results are representative of three similar experiments. (C) BLI analyses of the sequential binding of HLE-nano-BiKEs, CD16 and albumin to immobilized CD38. The biotinylated recombinant ectodomain of human CD38 was bound to a streptavidin-coated BLI sensor (30-90s). HLE-nano-BiKEs, recombinant ectodomain of human CD16, and human albumin were added at the times indicated. In each case binding was measured for 60s with a subsequent 60s dissociation phase. Arrows indicate the time point of protein addition; dashed lines indicate the start of the dissociation phase.