Reactive oxygen species induced by therapeutic CD20 antibodies inhibit natural killer cell-mediated antibody-dependent cellular cytotoxicity against primary CLL cells

Abstract

The antibody-dependent cellular cytotoxicity (ADCC) of natural killer (NK) cells is assumed to contribute to the clinical efficacy of monoclonal antibodies (mAbs) in chronic lymphocytic leukemia (CLL) and other hematopoietic malignancies of B cell origin. We sought to determine whether reactive oxygen species (ROS)-producing monocytes regulate the ADCC of NK cells against primary CLL cells using anti-CD20 as the linking antibody. The monoclonal CD20 antibodies rituximab and ofatumumab were found to trigger substantial release of ROS from monocytes. Antibody-exposed monocytes induced NK cell apoptosis and restricted NK cell-mediated ADCC against autologous CLL cells. The presence of inhibitors of ROS formation and scavengers of ROS preserved NK cell viability and restored NK cell-mediated ADCC against primary CLL cells. We propose that limiting the antibody-induced induction of immunosuppressive ROS may improve the anti-leukemic efficacy of anti-CD20 therapy in CLL.

Keywords: NK cells; NOX2; immunotherapy; monoclonal antibodies; reactive oxygen species.

Figure 1. Rituximab and Ofatumumab triggered ROS production by monocytes

Monocytes derived from healthy blood donors were continuously assessed for extracellular ROS production by chemiluminescense in the presence of CLL cells (Mo:CLL ratio 2:1) and the presence or absence of CD20 mAbs (10μg/ml). A. shows a representative kinetic graph, while the inset column diagram shows total ROS production (Area Under Curve; AUC; n = 4). B. shows total ROS production in presence and absence of OFA (10μg/ml) and OFA-derived F(ab')₂ fragments (10μg/ml). C., D. Total ROS production in presence and absence of the ROS formation inhibitors histamine dihydrochloride (HDC;100μM) and diphenylene iodonium chloride (DPI; 3μM) (n = 4). Statistical significance for all figures was determined by one-way ANOVAs and the Bonferroni post test. (RLU;relative light units) *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2. Monocytes restricted NK cell ADCC against autologous leukemic cells by production of ROS

A., B. NK cells and CFSE-labeled CLL cells were co-cultured for four hours in the presence or absence of autologous monocytes at an NK:Mo:CLL-ratio of 2:2:1 and IL-2 (500IU/ml), rituximab (10μg/ml), HDC (100μM), ranitidine (Ran; 100μM) or catalase (Cat; 200IU/ml). ADCC was inhibited by the presence of monocytes, but largely restored by anti-oxidative agents HDC or catalase. (n = 5-7). C. Representative dot-plot depicting the read-out for lysed leukemic cells of panels A and B. Percentages denote the proportion of lysed leukemic cells, thus staining positive for the Live/Dead stain. D. Monocytes were found to decrease the density of surface-bound rituximab on CLL cells, a mechanism referred to as trogocytosis. E. NK cell-mediated ADCC of CLL cells previously exposed to monocytes, and thus allowing for antigen removal by trogocytosis, was lowered in 7 out of 8 performed experiments. F. Monocyte-mediated trogocytosis was unaffected by addition of anti-oxidative substances (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3. CD20 antibodies induced ROS-dependent NK cell apoptosis

A. NK cells and monocytes were co-cultured for 18 hours at a Mo:NK-ratio of 1:2 in presence or absence of immobilized, plate-bound RTX (5μg/ml; left) or OFA (5μg/ml; right) and HDC (100μM), catalase (200IU/ml) or DPI (3μM) (n = 4). B. Representative dot-plot of the read-out for dead NK cells shown in panel A. C. PBMCs and PMNs at equal ratios were cultured for 18 hours in the presence or absence of plate-bound RTX, OFA or OFA- derived F(ab’)₂ (10 μg/ml) and DPI (3μM). Bars show percentage of remaining, viable CD 56⁺CD3⁻ NK cells compared to cultures with no antibody added (dotted line; n = 5). ***p < 0.001.