Anti-tumor activity of obinutuzumab and rituximab in a follicular lymphoma 3D model

Abstract

Follicular lymphomas (FLs) account for 35-40% of all adult lymphomas. Treatment typically involves chemotherapy combined with the anti-CD20 monoclonal antibody (MAb) rituximab (RTX). The development of the type II anti-CD20 MAb obinutuzumab (GA101) aims to further improve treatment. Here, using FL cells we show that RTX and GA101 display a similar activity on RL cells cultured in 2D. However, 2D culture cannot mimic tumor spatial organization and conventional 2D models may not reflect the effects of antibodies as they occur in vivo. Thus, we created a non-Hodgkin's lymphoma (NHL) 3D culture system, termed multicellular aggregates of lymphoma cells (MALC), and used it to compare RTX and GA101 activity. Our results show that both antibodies display greater activity towards FL cells in 3D culture compared with 2D culture. Moreover, we observed that in the 3D model GA101 was more effective than RTX both in inhibiting MALC growth through induction of (lysosomal) cell death and senescence and in inhibiting intracellular signaling pathways, such as mammalian target of rapamycin, Akt, PLCgamma (Phospholipase C gamma) and Syk. Altogether, our study demonstrates that spatial organization strongly influences the response to antibody treatment, supporting the use of 3D models for the testing of therapeutic agents in NHL.

Figure 1

Effect of RTX and GA101 on 2D-cultured RL cells. (a) Cell viability of RL cells treated or not with 10 μg/ml of RTX or GA101 was analyzed by exclusion of Trypan blue. (b) Percentage of cells in early and late apoptosis (Annexin V⁺/7AAD⁻ and Annexin V⁺/7AAD⁺) were determined by flow cytometry after 24, 48 and 72 h of treatment. Results are mean ±s.d. of at least three independent experiments. *P<0.05 compared with UT cells.

Figure 2

Effect of RTX and GA101 on MALC. (A) Pictures of treated or UT MALC were taken with an inverted Nikon Eclipse TE200 microscope at magnification × 40 at different times of culture with or without Hoeschst 33342 staining. These pictures are representative of several experiments (d=day). (B) MALC were treated or not (▪) with 10 μg/ml RTX (▴) or GA101 (●). MALC volume was measured at different times of culture. Results are expressed in mm³ and represent the mean±s.d. of nine independent experiments. Asterisk represents significant differences between UT and anti-CD20 MAb treatment (P<0.05); double asterisks represents significant differences between RTX and GA101 treatment (P<0.05); (d=day). (C) Number of viable cells per treated or UT MALC over time. Histograms represent mean±s.d. of nine independent experiments. Asterisk represents significant differences between UT and anti-CD20 MAb treatment (P<0.05); double asterisks represents significant differences between RTX and GA101 treatment (P<0.05); (d=day). (D) Left panel: necrosis was determined by flow cytometry analyzing Annexin V⁺/7AAD⁺ cells in MALC at 5, 10 and 20 days. Right panel: Ki67 immunohistochemistry labelling in MALC at day 10. (Ea) Fibronectin, laminin and vitronectin expression analyzed by western blotting in MALC compared with 2D RL cells. β-Actin expression was used as a control of protein expression. (Eb) Fibronectin and laminin labelling on paraffin sections of RL xenografts or lymph nodes isolated from FL patients. (Ec) Collagen I (yellow arrows) was visualized by second-harmonic generation in a lymph node from FL patients (left) and MALC (right). Results are representative of three independent experiments. (F) TV and tumor weight were measured after RL engraftment onto SCID-Beige mice. Histograms represent the mean±s.d. of 9 animals for PBS, 9 animals for RTX and 10 animals for the GA101-treated group. Double asterisks represents significant differences between RTX and GA101 treatment (P<0.05).

Figure 3

Induction of MALC apoptosis by RTX and GA101. (a) MALC were treated or not with 10 μg/ml of Trastuzumab (TTZ), RTX or GA101 for 5–20 days (d). Early apoptotic cells (Annexin V⁺/7AAD⁻) were detected by flow cytometry. Dot plots are representative of one experiment, and histograms represent the mean ±s.d. of FIVE independent experiments. Asterisk represents significant differences between UT and anti-CD20 MAb treatment (P<0.05). (b) Detection of PARP and caspase 3 by western blot analysis in MALC treated or not with 10 μg/ml of TTZ, RTX or GA101 for 5–20 days (d). Results are representative of three independent experiments. β-Actin expression was used as a control of protein expression. (c) Caspase 3 detection by confocal miscroscopy in fixed and embedded representative MALC treated or not with 10 μg/ml of RTX or GA101 for 20 days. Magnification × 40. (d) Detection of cytochrome c release from mitochondria using flow cytometry in MALC treated or not with 10 μg/ml of TTZ, RTX or GA101 for 20 days. Histograms were analyzed with Cytobank (www.cytobank.org) and are representative of three independent experiments. (e) Mitochondrial depolarization was analyzed by flow cytometry in MALC treated (10 μg/ml antibody) or not for 20 days. Histograms represent the mean percentage of red-JC-1-negative cells for four independent experiments ±s.d. Asterisk represents significant differences between UT and anti-CD20 MAb treatment (P<0.05). (f) ROS production was measured in MALC treated or not for 20 days with 10 μg/ml TTZ, RTX or GA101 by flow cytometry. Cells producing ROS were analyzed by double staining with Mitotracker green⁺/Mitotracker deep red⁻, and results represent the mean percentage of four independent experiments ±s.d. Asterisk represents significant differences between UT and anti-CD20 MAb treatment (P<0.05). (g) Active caspase 3 quantification was performed on tumors derived from FL xenograft SCID-Beige mice injected intraperitoneally with PBS or with antibodies at 25 mg/kg twice a week.

Figure 4

GA101 induces lysosomal cell death (a) and senescence (b) in MALC. (a) Left panel: Detection of lysosomal membrane permeabilization after acridine orange (AO) staining of MALC treated or not with 10 μg/ml antibodies for 20 days. Box shows overlay of fold changes compared with UT MALC for one representative experiment. Histograms represent the mean fold change ±s.d. of five independent experiments. *P<0.05. Analysis and representation were performed with Cytobank. Insert: 2D RL culture cells treated with 50 μℳ chloroquine or 50 nℳ bafilomycin for 3 or 24 h were used as positive controls for lysosomal cell death. Right panel: MALC were treated (10 μg/ml antibodies) or not for 20 days, and cathepsin D expression was determined by western blot analysis. Results are representative of three independent experiments. β-Actin expression was used as a control of protein expression. (b) Detection of cleavage of C₁₂FDG by SA-βGal was measured by an increase in green fluorescence in treated and UT MALC after 20 days of culture. 2D RL culture cells treated with 2.2 μℳ cisplatin or 0.9 μℳ etoposide for 3 days were used as a positive control of senescence. Left: Overlays of fold changes comparing treated to UT cells in one representative experiment. Right: Histograms represent the mean fold change ±s.d. of five independent experiments *P<0.05. Analysis and representation were performed with Cytobank.

Figure 5

GA101 affects MALC signaling pathways. Cell signaling was analyzed as described in Materials and Methods. Dot plots represent all encoded conditions discriminated according to the CBD450 fluorescence of either 2D RL cells treated for 72 h with 10 μg/ml antibody (a) or MALC after 20 days of antibody treatment (b). Phosphospecific stainings were analyzed for each gated condition and heatmaps represent relative levels of phospho-Akt/Akt, phospho-Syk/Syk and phospho-PLCγ2/PLCγ2 for one representative experiment in RL treated or not for 72 h or in MALC treated or not at day 20. Analysis and representation were performed with Cytobank. Histograms represent the relative percentage of phosphorylated proteins in treated 2D (a) or 3D (b) cells compared with UT cells. Results represent mean±s.d. of three independent experiments *P<0.05. Western blot analysis of p70S6K phosphorylation was performed in 2D RL cells treated or not with 10 μg/ml antibody for 72 h (a).

Figure 6

Antibodies sensitize MALC to rapamycin. MALC were treated in the drop (a) or after 5 days of culture (b) with 10 μg/ml antibodies. Twenty-four hours later, MALC were treated with 50 nℳ of rapamycin. Every 5 days, cells were submitted to the same treatment. Results represent the percentage of cells in sub-G1 phase as determined by flow cytometry after DAPI staining at day 20 and are the mean of six independent experiments ±s.d. Asterisk represents significant differences between UT and anti-CD20 MAb/rapamycin treatment (P<0.05); Double asterisks represents significant differences between RTX/Rapa and GA101/Rapa treatment (P<0.05).