Apo2L/TRAIL inhibits tumor growth and bone destruction in a murine model of multiple myeloma

Abstract

Purpose: Multiple myeloma is an incurable disease, for which the development of new therapeutic approaches is required. Here, we report on the efficacy of recombinant soluble Apo2L/tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) to inhibit tumor progression and bone destruction in a xenogeneic model of human multiple myeloma.

Experimental design: We established a mouse model of myeloma, in which Apo2L/TRAIL-sensitive RPMI-8226 or KMS-11 cells, tagged with a triple reporter gene construct (NES-HSV-TK/GFP/Luc), were transplanted directly into the tibial marrow cavity of nude mice. Tumor burden was monitored progressively by bioluminescence imaging and the development of myeloma-induced osteolysis was measured using high resolution in vivo micro-computed tomography.

Results: Tumor burden increased progressively in the tibial marrow cavity of mice transplanted with Apo2L/TRAIL-sensitive RPMI-8226 or KMS-11 cells associated with extensive osteolysis directly in the area of cancer cell transplantation. Treatment of mice with recombinant soluble Apo2L/TRAIL reduced myeloma burden in the bone marrow cavity and significantly protected against myeloma-induced osteolysis. The protective effects of Apo2L/TRAIL treatment on bone were mediated by the direct apoptotic actions of Apo2L/TRAIL on myeloma cells within the bone microenvironment.

Conclusions: This is the first in vivo study that investigates the efficacy of recombinant Apo2L/TRAIL on myeloma burden within the bone microenvironment and associated myeloma-induced bone destruction. Our findings that recombinant soluble Apo2L/TRAIL reduces myeloma burden within the bone microenvironment and protects the bone from myeloma-induced bone destruction argue against an inhibitory role of osteoprotegerin in Apo2L/TRAIL-induced apoptosis in vivo and highlight the need to clinically evaluate Apo2L/TRAIL in patients with multiple myeloma.

Fig. 1

Apo2L/TRAIL-induced apoptosis in multiple myeloma cell in vitro. A, comparison of the effect of Apo2L/TRAIL in a panel of well-established human multiple myeloma cell lines. Cells were left untreated or treated with increasing Apo2L/TRAIL concentrations of 3,10, 30,100, and 100 ng/mL and cell viability was determined by MTTassay 24 h thereafter. Representative experiment repeated at least three times. Points, mean; bars, SD. B, cells were seeded on chamber slides at 5 × 10⁴ per chamber and treated with Apo2L/TRAIL at 100 ng/mL for 24 h. Cells were fixed with methanol and incubated with 4′,6-diamidine-2-phenylindole before washing in PBS and mounting on PBS/glycerin. 4′,6-Diamidine-2-phenylindole staining was visualized by fluorescence microscopy. C, left, RPMI-8226 cells were treated with increasing concentrations of Apo2L/TRAIL for 24 h. Cell lysates were used to determine caspase-3-like activity using the caspase-3 specific fluorogenic substrate, zDEVD-AFC, as described in Materials and Methods. Right, RPMI-8226 cells were treated for 24 h with 100 ng/mL Apo2L/TRAIL alone or coincubated with the broad-specificity caspase inhibitor zVAD-fmk (50 μmol/L). To exclude possible toxic effects of the inhibitor, cells were also treated with the inhibitor alone. Cell viability was determined using the MTTassay and expressed as percentage of control. Quadruplicate results from a representative experiment repeated at least twice. Points, mean; bars, SD. D, RPMI-8226 cells were seeded at 2 × 10⁶ perT25 flask and either left untreated or treated with Apo2L/TRAIL at a concentration of 100 ng/mL. Cells were then lysed and total cell lysates were analyzed by PAGE and transferred to polyvinylidene fluoride membranes for immunodetection. The caspase-8, caspase-9, caspase-10, and poly(ADP-ribose) polymerase antibodies detect both full-length and processed forms of the antigen, whereas caspase-3and Bid antibodies detect only the full-length antigens. E, comparison of the effect of Apo2L/TRAIL on NES-HSV-TK/GFP/Luc retrovirally infected cells (RPMI-8226-NES-TGL) and their parental noninfected counterpart cells (RPMI-8226). Cells were left untreated or treated for 24 h with increasing concentrations Apo2L/TRAIL. Cell viability (top) and caspase-3 activity (bottom) were determined as described above. Representative experiments repeated at least three times. Bars, SD.

Fig. 2

Noninvasive in vivo BLI and quantification of the effect of Apo2L/TRAIL on multiple myeloma growth in the bone marrow. A, nude mice were injected with 1 × 10⁶ RPMI-8226-NES-TGL cells into the left tibia, whereas the right contralateral tibia was injected with PBS alone as an internal control. Two independent experiments were carried out. For each experiment, mice were randomized into two groups of 10 mice per group. Seven days after cancer cell transplantation, Apo2L/TRAIL was administered (intraperitoneally) at 30 mg/kg/dose once per day for 5 consecutive days followed by a once weekly dose until termination of the experiment. Vehicle-treated animals received intraperitoneal injections of PBS adhering to the same schedule of Apo2L/TRAIL administration. B, nude mice (n = 6) were injected with 1 × 10⁶ KMS-11-NES-TGL cells into the left tibia and tumors were allowed to progress to a define size as measured by photon counts per second. Apo2L/TRAIL was administered on day 21 after cancer cell transplantation once per day for 5 consecutive days followed by a once weekly dose. Mice were imaged weekly using the Xenogen IVIS 100 BLI system. Representative whole-body images of untreated and treated animals during the course of the experiments. BLI measurements are expressed as the sum of integrated photon counts per second. Mean ± SE. A significant difference between the vehicle group and the Apo2l/TRAIL-treated group was observed on days 28 and 35 (*, P < 0.01) in experiment 1 and on day 35 in experiment 2 (*, P < 0.05).

Fig. 3

Noninvasive in vivo micro-CT imaging and quantification of the effect of Apo2L/TRAIL on multiple myeloma-associated bone osteolysis. A, representative three-dimensional reconstructions of micro-CT images generated from serial two-dimensional cross-sectional micro-CT images (18 μm resolution) covering the total length of each tumor-bearing tibiae taken on days 14, 28, and 42 after cancer cell transplantation. Progression of myeloma-induced osteolysis in the animals over time using in vivo micro-CT Treatment with Apo2L/TRAIL resulted in a significant conservation of the tibiae concomitant with the reduction in myeloma burden in the marrow as assessed by BLI. B, bone morphometric parameters from tibiae of vehicle-treated (n = 8) or Apo2L/TRAIL-treated (n = 8) animals were quantified at the time of sacrifice using ex vivo high-resolution two-dimensional micro-CT images (5 μm resolution) as described in Materials and Methods. The effect of Apo2L/TRAIL on total bone volume (Total BV) and trabecular bone volume (Trab BV) is expressed as the percentage reduction relative to mean value of vehicle-treated mice at end of the experiment (*, P < 0.05). C, representative H&E-stained decalcified tibial sections showing intense tunnel positive staining confirming the presence of apoptosis in a substantial fraction of the tumor (T) in the Apo2L/TRAIL-treated animals when compared with the tumor-bearing vehicle-treated animals or contralateral non-tumor-bearing tibia of Apo2L/TRAIL-treated animals, respectively.

Fig. 4

Development of Apo2L/TRAIL resistance with prolonged treatment in vivo and in vitro. A, at the end of the experiment (day 42), RPMI-8226-NES-TGL cells were flushed from the tibiae of vehicle-treated (RPMI-8226-NES-TGL-vehicle) or Apo2L/TRAIL-treated (RPMI-8226-NES-TGL-Apo2L/TRAIL) animals and cultured ex vivo for 24 h in the presence of 100 ng/mL Apo2L/TRAIL. Cell viability was assessed by the MTTassay (left), whereas caspase-3activity was measured from cell extracts isolated from the same cells as described in Materials and Methods (right). Solid columns, RPMI-8226-NES-TGL-vehicle cells; empty columns, RPMI-8226-NES-TGL-Apo2L/TRAIL. Average ± SD of triplicates. P < 0.05. B, Apo2L/TRAIL-resistant cells were also generated in vitro by culturing the parental RPMI-8226-NES-TGL cells continuously in medium containing Apo2L/TRAIL for ∼8wkas described in Materials and Methods. These resistant cells (denoted RPMI-8226-NES-TGL-R), when compared with the parental cells (RPMI-8226-NES-TGL), were completely resistant toApo2L/TRAIL, showing ∼100% viability even at the highest dose of 100 ng/mL Apo2L/TRAIL (left) and showing lack of Apo2L/TRAIL-induced caspase-3activity (right). C, flow cytometric analysis for cell surface expression of Apo2L/TRAIL receptors in the parental sensitive RPMI-8226-NES-TGL cells (dark solid line), resistant RPMI-8226-NES-TGL-R cells (dotted line), and intermediately sensitive RPMI-8226-NES-TGL-Apo2L/TRAIL cells (light solid line). Depicted graphs were obtained after staining with anti human TRAIL/DR4,TRAIL/DR5,TRAIL/DcR1, and TRAIL/DcR2 monoclonal antibodies as described in Materials and Methods. Shaded curves, isotype control staining with the IB5 IgG1 mAb; unshaded curves, staining with the respective Apo2L/TRAIL receptor antibodies.

Fig. 5

Qualitative and quantitative assessment of the effect of Apo2L/TRAIL treatment on normal bone. Representative three-dimensional reconstructed cross-sectional and longitudinal micro-CT images (5 μm resolution) of contralateral normal non-tumor-bearing tibiae from vehicle-treated and Apo2L/TRAIL-treated mice. The antiresorptive bisphosphonate zoledronic acid (ZOL) was used as a positive control for comparison in a separate cohort of animals (n = 10). Bone morphometric parameters including total bone volume and trabecular bone volume were assessed using micro-CT software as described in Materials and Methods. The extended evaluation of bone morphometric parameters is shown in Table 2. Data indicate that treatment with zoledronic acid (100 mg/kg/dose once weekly for 5 wk) resulted in significant changes in all bone morphometric parameters as expected. In contrast, treatment with Apo2L/TRAIL was without effect.