Novel type II anti-CD20 monoclonal antibody (GA101) evokes homotypic adhesion and actin-dependent, lysosome-mediated cell death in B-cell malignancies

Abstract

The anti-CD20 mAb rituximab has substantially improved the clinical outcome of patients with a wide range of B-cell malignancies. However, many patients relapse or fail to respond to rituximab, and thus there is intense investigation into the development of novel anti-CD20 mAbs with improved therapeutic efficacy. Although Fc-FcγR interactions appear to underlie much of the therapeutic success with rituximab, certain type II anti-CD20 mAbs efficiently induce programmed cell death (PCD), whereas rituximab-like type I anti-CD20 mAbs do not. Here, we show that the humanized, glycoengineered anti-CD20 mAb GA101 and derivatives harboring non-glycoengineered Fc regions are type II mAb that trigger nonapoptotic PCD in a range of B-lymphoma cell lines and primary B-cell malignancies. We demonstrate that GA101-induced cell death is dependent on actin reorganization, can be abrogated by inhibitors of actin polymerization, and is independent of BCL-2 overexpression and caspase activation. GA101-induced PCD is executed by lysosomes which disperse their contents into the cytoplasm and surrounding environment. Taken together, these findings reveal that GA101 is able to potently elicit actin-dependent, lysosomal cell death, which may potentially lead to improved clearance of B-cell malignancies in vivo.

Figure 1

GA101 induces Fc-independent PCD and HA. (A) A panel of CD20-positive, human B-lymphoma cell lines was treated with 10 μg/mL mAb and cell death was analyzed after 24 hours using the annexin V–FITC/PI assay. The anti-HER2 mAb trastuzumab was used as a human IgG1 isotype control. GA101 induced significantly higher cell death than rituximab in all the cell lines tested (for Raji and Granta 519, P < .001; Daudi, P < .01; SU-DHL4, P < .03). (B) Raji cells were treated with GA101 (10 μg/mL) for 4 hours and the viable cell population was isolated, rested overnight, then re-treated with GA101 for 4 hours. Subsequently, cell death was analyzed as in panel A. Mean and SEM of 2 independent experiments are shown. Cell death induced in viable isolated cells was equivalent to that induced in previously untreated cells. (C) The extent of HA induced by mAbs (10 μg/mL) was assessed by low-magnification light microscopy 4 hours after treatment, and HA in Daudi cells is shown as an example (scale bar, 500 μm). GA101 induced superior HA than rituximab. (D) Cell death induced by GA101 in Raji cells was directly compared with that induced by a non-glycoengineered, wild-type human IgG1-bearing derivative of GA101, GA101 (NG), and F(ab)′₂ fragments of GA101. All induced equivalent amounts of cell death confirming that GA101-induced cell death is Fc independent. (E) ⁵¹Cr-release cell death assay. Cells were prelabeled with ⁵¹Cr for 1 hour at 37°C before treatment with mAb and after 4 hours, ⁵¹Cr release was measured as described in “Cell death and cell viability assays.” Mean + SEM of quintuplicate samples representative of 2 independent experiments are shown. GA101 induced significantly greater ⁵¹Cr release than rituximab, and F(ab)′₂ fragments of GA101 were sufficient for the induction of ⁵¹Cr release.

Figure 2

Cell death and HA induced by GA101 is dependent on actin polymerization. (A) Time-lapse microscopy of Raji cells following treatment with GA101. Cells were suspended in phenol red-free media containing 7-aminoactinomycin D (7-AAD, 5 μg/mL) as a marker for cell death, and treated with GA101 (10 μg/mL) at time 0. Phase-contrast and fluorescence images were captured every 5 minutes and overlaid, with a sample of images captured at different time points shown (scale bar, 20 μm). Cellular adhesion was followed by cell swelling, loss of plasma membrane integrity, and cell death. Arrow marks a morphological control for apoptosis shown on the top left of the image. (B) Morphological pattern of cell death induced by GA101. Images show adhering cells showing gross cytoplasmic disintegration and loss of plasma membrane integrity after treatment with GA101. Staurosporine (STSP) was used as a morphological control for apoptosis. (C) Cells were incubated with inhibitors of actin polymerization (cytochalasin D and latrunculin B, 10μM) before treatment with mAb. Cell death and homotypic adhesion were analyzed 4 hours after treatment. Disruption of the actin cytoskeleton significantly inhibited cell death induced by GA101 and HA as shown in the example with latrunculin B (scale bar, 100 μm; *P < .02 **P < .01).

Figure 3

GA101 induces actin-dependent PCD and HA in primary B-CLL. (A) Primary B-CLL cells were isolated from patient blood samples as described in “Primary tumor samples,” preincubated with vehicle control (DMSO) or latrunculin B (Lat B, 10μM) then treated with anti-CD20 mAbs (5 μg/mL) for 4 hours, and cell death was assessed using annexin V-Cy5.5/7-AAD staining. Mean ± SEM of 4 independent patient samples are shown. Because of the heterogeneous levels of background cell death in individual patient samples, death was expressed as percentage above control. GA101 and GA101 (NG), the non-glycoengineered derivative of GA101, both induced significantly higher cell death than rituximab and ofatumumab. Cell death was completely ablated by latrunculin B (*P < .005, **P < .0001). (B) Primary B-CLL cells were treated as above and assessed for HA using light microscopy. Figure shows representative images from patient CLL31 (scale bar, 50 μm). GA101-induced HA correlates with PCD, and is blocked by the disruption of the actin cytoskeleton with latrunculin B.

Figure 4

GA101-induced cell death is independent of DNA fragmentation, caspase activation, and BCL-2 overexpression. (A) Raji cells were treated with anti-CD20 mAb (10 μg/mL) or staurosporine (STSP, 2μM), a positive control for apoptosis, for 24 hours and DNA fragmentation was assessed using TUNEL staining analyzed by flow cytometry. Mean ± SEM of 3 independent experiments is shown on the left, with representative plots on the right. GA101 does not induce significant DNA fragmentation. (B) Wild-type Raji (left) and Raji cells that overexpress the antiapoptotic protein BCL-2, (Raji-BCL2; right), were preincubated with DMSO or Q-VD-OPH (20μM) for 30 minutes, then treated with anti-CD20 mAbs (10 μg/mL) or mitoxantrone (1 μg/mL), and cell death measured 48 hours later. Mean + SEM of 4 independent experiments are shown. Neither BCL-2 overexpression, caspase inhibition, nor a combination of both, had any impact on cell death induced by GA101, despite inhibiting mitoxantrone-induced apoptosis (P < .008). (C) Raji and Raji-BCL2 cells were treated as in panel B and growth inhibition assessed using the XTT assay as described in “Cell death and cell viability assays” 48 hours after treatment, with absorbance normalized relative to untreated cells. Mean + SEM of 4 independent experiments are shown. Caspase inhibition and BCL-2 overexpression had no impact on GA101-induced growth inhibition, both of which significantly attenuated growth inhibition induced by chemotherapy (mitoxantrone and doxorubicin, 1 μg/mL; P < .01). (D) Raji cells were treated with anti-CD20 mAbs or mitoxantrone as described above and Western blot analysis was performed for cleaved caspase 3 (CC3). No CC3 was observed following anti-CD20 mAb treatment (Ctl indicates control mAb; GA, GA101; Rit, rituximab; and Mtx, mitoxantrone).

Figure 5

Lysosomal changes associated with GA101-induced cell death. (A) To determine changes in lysosomal volume, Raji cells were treated with mAbs (10 μg/mL) for 1 or 4 hours, labeled with Lysotracker green (75nM) and analyzed by flow cytometry. Histograms represent cells treated with control mAb (black), GA101 (blue), rituximab (purple), and unlabeled cells to set the background (solid red). GA101 induced an enlargement of the lysosomal compartment at 1 hour and a subsequent collapse in a subpopulation of the cells at 4 hours, whereas rituximab induced no changes in lysosomal volume. (B) To directly correlate cell death with lysosomal volume, cells were treated with mAbs. After 24 hours, cells were costained with Lysotracker green and annexin V Cy5.5 to label the dead cell population and analyzed by flow cytometry. Staurosporine (STSP) was used as a positive control for apoptosis. Cell death evoked by GA101 was associated with a collapse of the lysosomal compartment (upper left quadrant, in red). (C) Cells were preincubated with the V-ATPase inhibitor concanamycin A (CMA, 100nM) before treatment with mAbs and death was analyzed 24 hours after treatment. CMA significantly inhibited cell death induced by GA101 (*P < .001). (D) Cells were treated as above and lysosomal volume was quantified using Lysotracker green staining 1 hour after treatment. Histograms represent cells treated with control mAb (black), control mAb + CMA (purple), GA101 (blue), GA101 + CMA (green), and background (solid red). CMA prevents the increase in lysosomal volume induced by GA101.

Figure 6

The role of lysosomal membrane permeabilization and lysosomal cathepsins in GA101-induced cell death. (A) Raji cells were incubated with acridine orange (AO) to label lysosomes, washed twice, treated with mAbs, and analyzed at different time points. Leakage of lysosomal contents into the cytoplasm was measured as an increase in green fluorescence detected by FACS. GA101 induced an increase in green fluorescence at 2 hours, followed by a subsequent loss of fluorescence at 4 and 8 hours. (B) Fluorescence microscopy of the lysosomal protease cathepsin B staining (red) of Raji cells 4 hours after treatment with mAbs. DNA was counterstained with DAPI (blue; scale bar, 40 μm). GA101 induces marked cathepsin B release into the cytosol and surrounding points of cellular adhesion. (C) Western blot of the active 25-kDa subunit of cathepsin B (CTSB) in cell supernatants 4 hours after treatment with mAbs. BSA was used as a loading control on the same supernatants diluted 1:5. GA101 evokes extracellular cathepsin B release (Ctl indicates control mAb; GA, GA101; and Rit, rituximab). (D) Cells were preincubated with cathepsin inhibitor III (100μM) before treatment with mAbs and cell death was analyzed 4 hours after treatment. Figure shows mean + SEM of triplicates, representative of 2 independent experiments. Cathepsin inhibitor III inhibits GA101-induced cell death (*P < .001).