Background
Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) was first identified in 1996 [
1]. TRAIL induces caspase-dependent apoptosis in cancer cells after binding to its corresponding receptors: DR4 or DR5 in humans and DR5 in rodents. Due to its cancer cell specific death-inducing activity, TRAIL has been regarded as a potential protector or inhibitor of tumor generation and progression [
2,
3]; however, lots of tumor cell lines and primary cancer cells are resistant to TRAIL-induced apoptosis [
4‐
6]. Additionally, TRAIL activates non-apoptotic signals, including NF-kB, mitogen activated kinase (MAPK), JNK, p38, Akt, and ERK [
7‐
13]. TRAIL actually induces the proliferation of vascular endothelial cells and specific cancer cells, rather than inhibiting them [
14‐
16]. These potential adverse effects accelerate research interest regarding the resistance mechanism of TRAIL-induced apoptosis and in the potential to combine TRAIL with an additional agent to induce apoptosis [
17‐
22].
Research to induce cell death in TRAIL resistant cells has demonstrated some success. TRAIL induced apoptosis has been achieved through combined treatment with certain chemical agents, as well as through the genetic modification and subsequent expression of certain apoptosis related proteins [
4,
7,
17,
18,
23‐
33]. Though regulating specific apoptosis-related proteins can induce cell death, this process is not applicable to all TRAIL-resistant cells because the resistance mechanisms are heterogeneous. According to published reports, several additional agents can be employed to induce apoptosis via TRAIL. Some agents regulate the balance between pro-apoptotic and anti-apoptotic proteins, while others modify membrane lipid composition to provide more opportunities for DR4 or DR5 to localize to membrane lipid rafts in specific cells. [
26,
34,
35]. Many researchers have found that apoptosis-related proteins involved in the TRAIL signaling pathway do not carry any mutations. Both the expression and membrane localizations of TRAIL receptors are well regulated in almost all TRAIL resistant cells.
In this study, we hypothesized that specific intracellular conditions would be required to induce cell death in TRAIL resistant cancer cells and that these conditions could be provided by an additional agent, such as bortezomib. In addition, the effective cell death inducing ability of recombinant TRAIL would be essential. We tested various recombinant TRAIL proteins to compare cell death inducing ability; either single treatment in TRAIL-susceptible cells, or combined treatment with bortezomib in TRAIL-resistant cells. Tumor progression was inhibited with the combined treatment of bortezomib and ILz:rhTRAIL; the latter being a recombinant human TRAIL protein containing isoleucine zipper hexamerization motif for efficient multimerization. The cell death induced by the combined treatment was found to be influenced by TRAIL blocking antibody and regulated by caspases. It was not affected by ER stress or autophagy.
Methods
Cell culture
CT26 (80009), B16F10 (80008), MDA-MB-231 (30026), and HEK 293 (21573) cells were purchased from the Korean Cell Line Bank (KCLB, Seoul, Korea). HeLa (HC18802) and Jurkat (HC18111) cells were purchased from the Korean Collection for Type Cultures (KCTC, Daejeon, Korea). 4 T1 (CRL-2539) cells were purchased from the American Type Culture Collection (ATCC). BMK (baby mouse kidney) cells were kindly provided by Dr. J. Hiscott. The cells were cultured, according to the recommendations of KCTC, in Gibco™ Dulbecco’s Modified Eagle Medium (DMEM) or RPMI-1640 (Thermo Fisher Scientific, Pittsburg, PA, USA) supplemented with 5% fetal bovine serum (Capricorn Scientific, Ebsdorfergrund, Germany) and 1% Gibco™ penicillin-streptomycin at 37 °C in a humidified atmosphere containing 5% CO2.
XTT assay
Cell viability was measured by XTT assay using Cell Titer 96® Aqueous one solution reagent (Promega, Madison, WI, USA), according to the manufacture’s protocol. Cells were seeded into a 96-well plate (1 × 104 cells/well) a day before treatment with recombinant TRAIL and/or bortezomib, which was purchased by Selleck Chemicals (Houston, TX, USA). The Medium was changed to phenol-red free DMEM or RPMI with 10% of FBS and then treated with recombinant TRAIL and/or bortezomib. After an appropriate time period, substrate solution in the kit was added into the medium, and optical density was measured at 490 nm using a microplate reader (Infinite M200™; Tecan, Männedorf, Switzerland). Data are presented as the relative percentage of viable cells using the average optical density (O.D.) of the untreated control as a reference (100%) in each experiment.
Preparation of cell lysates
Cell lysates were prepared for immunoblotting analysis. Cells were harvested with ice-cold RIPA buffer (50 mM Tris-HCl pH 7.4, 1% NP-40, 0.5% Na-deoxycholate, 0.1% SDS, 150 mM NaCl, 2 mM EDTA) and supplemented with phosphatase inhibitor cocktails-I, -II (Merck Millipore, Darmstadt, Germany). Proteins were quantified using Pierce™ BCA (bicinchoninic acid) protein assay kit (Thermo Fisher Scientific). Protein samples were loaded onto SDS-PAGE gel and transferred to Polyvinylidene difluoride (PVDF) membrane (Amersham, Buckinghamshire, England).
Immunoblotting analysis and antibodies
To examine expression levels of specific proteins, immunoblotting analysis was performed using appropriate antibodies. For immunoblotting, the membranes were immersed for 1 h into Tris-buffered saline (TBS), containing 5% skim milk (Thermo Scientific), then incubated with appropriate primary antibodies. After 3 h of incubation at room temperature or overnight at 4 °C, the membrane was washed with 0.05% TBST solution (TBS with 0.05% Tween-20) three times for 10 min each and incubated for 1 h with horseradish peroxidase-conjugated secondary antibody (Jackson ImmunoResearch, West Grove, PA, USA). After extensive washing, immunoreactive proteins were detected using PowerOpti-ECL™ solution (Animal Genetics, Suweon, Korea). Intensities of reactive bands were measured and compared using Image J software (
www.ImageJ.net).
Antibodies
Antibodies were used to neutralize TRAIL signaling and for immunoblotting analysis. Anti-TRAIL antibody (B-T24) for neutralization was purchased from Diaclone (Besançon cedex, France). For immunoblotting analysis, antibodies against caspase 3 (9662), caspase 9 (9508), Bcl-2 (2876), Bcl-XL (2764), and Mcl-1 (5453) were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-caspase-8 (ALX-804-477) antibody was purchased from Enzo Life Sciences (Farmingdale, NY, USA) and anti-β-actin antibody (MAB1501) was purchased from Millipore (Thermo Fisher Scientific). Anti-XIAP antibody (sc-55,551) was purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA).
Recombinant TRAIL proteins
Recombinant human and mouse TRAIL proteins (rhTRAIL and rmTRAIL) were purchased from PeproTech (PeproTech Korea, Seoul, South Korea): rhTRAIL (310–04); rmTRAIL (315–19). ILz:rhTRAIL was constructed and purified by Dr. Kim, as described in a previous report [
36]. Briefly, hexamerization isoleucine zipper is fused to the N-terminal of recombinant human TRAIL (amino acid 114–281) resulting in ILz:rhTRAIL. The recombinant ILz:rhTRAIL protein was expressed in
E. coli BL21 (DE3) and purified using a Ni-NTA His affinity column (Novagen, Carlsbad, CA, USA) according to the manufacturer’s instructions.
Flow cytometry
Cells were treated with ILz:rhTRAIL and bortezomib, with or without pre-treatment with z-VAD-fmk for 1 h. After 24 h of treatment with ILz:rhTRAIL and bortezomib, cells were harvested by trypsin treatment and stained with propidium iodide (PI) and Annexin V using the FITC Annexin V Apoptosis Detection Kit (BD Biosciences, Thermo Fisher Scientific) for 5 min. Stained cells were analyzed by BD FACS Calibur™ analyzer and the BD CellQuest™ program (BD Biosciences, Thermo Fisher Scientific).
Animal experiment
A syngeneic tumor model was generated to analyze whether the combined treatment of TRAIL with bortezomib could have an effect on tumor regression. Cultured CT26 cells were harvested and re-suspended with phosphate-buffered saline (PBS) following trypsin treatment. Sixty Balb/c mice, provided by Orient Bio (Sungnam, Korea), were divided into 4 groups: ‘sham’, ‘bortezomib’, ‘TRAIL’, and ‘TRAIL and bortezomib’. Cells were subcutaneously injected into the backs of all seven-week old Balb/c mice (2 × 105 cells/mouse), except for those mice in the ‘sham’ group.The size of each tumor was measured using a caliper (CD-15CPX, Mitutoyo, Japan) and tumor volume was calculated according to the equation: tumor volume = (L × W × W)/2, L = length, W = width. Ten days following the subcutaneous injection of cells, when average tumor volumes around 40 mm3 were reached, ILz:rhTRAIL (10 μg/kg) and/or bortezomib (3.8 μg/kg) were intravenously injected via tail vein. This consisted of a series of 5 injections, with a two-day interval between each injection. After these 5 injections were completed, tumor tissues were harvested, fixed with 10% formalin solution, and embedded with paraffin. Animal experiments were performed at Chosun University in accordance with the guidance of Chosun University Institutional Animal Care and Use Committee (acceptance number: CIACUC2016-A0023).
Statistically analysis
Statistical significance was determined by Student’s t-test or ANOVA single test. A two-tailed p < 0.05 was considered statistically significant.
Discussion
In this report, we used three kinds of recombinant soluble TRAIL proteins to identify cell death inducing ability. Two of the recombinant TRAIL proteins were from PeproTech Korea, one made with extracellular domains of mouse TRAIL sequences (rmTRAIL), while the other was composed of human TRAIL sequences (rhTRAIL). The third, ILz:rhTRAIL, contained an extracellular domain of recombinant human TRAIL and an isoleucine zipper hexamerization motif. ILz:rhTRAIL demonstrated the strongest cell death inducing ability (Fig.
1). This was seen both in TRAIL susceptible cells, and in TRAIL resistant cells treated with the bortezomib (Fig.
2). Using these recombinant TRAIL proteins, we investigated TRAIL-induced apoptosis by single treatment with TRAIL and by the combined treatment with TRAIL and bortezomib in both TRAIL susceptible and TRAIL-resistant human and murine cells. Multiple cell types were chosen based on their susceptibility to TRAIL-induced apoptosis and their species origin. Some researchers have theorized that the TRAIL susceptibility of murine cells may be dependent upon the species origin of the TRAIL proteins. However, in this report, death in TRAIL-resistant murine cells was induced by combined treatment with ILz:rhTRAIL and bortezomib. To investigate the cell death-inducing ability of the combined treatment in transformed cells, HEK 293 and BMK cells were used. HEK293 cells are transformed human cells susceptible to TRAIL-induced apoptosis, whereas BMK cells are transformed cells resistant to TRAIL-induced apoptosis.
Although four TRAIL-resistant cells (CT26, B16F10, BMK and 4 T1) were shown to be resistant or moderately resistant to bortezomib, these cells demonstrated significant susceptibility to bortezomib when combined with recombinant TRAIL proteins. Though ILz:rhTRAIL contains human TRAIL sequences, when compared to the other recombinant TRAIL proteins used in this report, ILz:rhTRAIL showed the highest cell death inducing ability in all cell types regardless of TRAIL susceptibility and species specificity. In the combination treatment, rmTRAIL showed stronger cell death inducing ability than rhTRAIL in all three murine cell lines (CT26, B16F10 and 4 T1), but not in BMK cells. Although species specific binding of TRAIL could affect the combined treatment-induced cell death in some murine cells, the structural characteristics, rather than species specific sequences, of recombinant TRAIL must be a major determining factor in their susceptibility to TRAIL-induced cell death with the combined treatment.
The combination treatment demonstrates potential for use in in vivo tumor systems, as regression was seen in 70% of syngeneic mouse tumors when treated with equivalent amounts of ILz:rhTRAIL (10 μg/kg mouse) and bortezomib (3.8 μg/kg mouse): 100 ng/ml of ILz:rhTRAIL; 100 nM of bortezomib (Fig.
3). Because 25 nM of bortezomib and 50 ng/ml of ILz:rhTRAIL were enough to induce CT26 cell death in Fig.
2b, dose reduction of ILz:rhTRAIL and bortezomib could produce similar tumor regression effects in syngeneic tumor models. To induce specific cancer cell death efficiently, further studies concerning the the development of TRAIL responsiveness, the dose reduction of bortezomib, and the development of bortezomib substitutes for use in the combined treatment would be necessary.
The expression modulations, generally observed in TRAIL-induced cell death, were detected in the combined treatment, but not in single treatment with either TRAIL or bortezomib. Considering that cell death rates were dependent on the characteristics of recombinant TRAIL, and that the cell death observed was caspase-dependent apoptosis, the combined treatment-induced cell death could be regulated by recombinant TRAIL in the intracellular conditions provided by the proteasome inhibitor. This suggestion is supported by the result that the combined treatment-induced cell death was inhibited by pre-incubation of anti-TRAIL antibody for blocking in CT26 and B16F10 cells (Fig.
4a). Although bortezomib can regulate various intracellular signaling processes involved in cell death, these were not deemed to be the direct cause of cell death in the combined treatment. Bortezomib induced the ER stress response and modulated autophagy in cells, however, ER stress inhibition and regulation of autophagy were not shown to modulate the cell death induced by the combined treatment of ILz:rhTRAIL and bortezomib (Additional file
5: Figure S5 and Additional file
6: Figure S6).
In the experiment to investigate whether cell death signaling could be transmitted by JNK activation, cell death appeared to be regulated by SP600125 in ILz:rhTRAIL treated cells based on the comparison of optical densities by XTT assay (Additional file
7: Figure S7 b). Although the optical density of the XTT assay in ILz:rhTRAIL pre-treated with SP600125 was reduced to the optical density seen in the combined treated cells, microscopic observation demonstrated no cell death (Additional file
7: Figure S7 c). This may be due to inhibition of the reactivity of the XTT substrate by SP600125. The ER stress response and JNK activation have been reported as either positive or negative regulators of TRAIL-induced cell death in several cells; however, the combination treatment-induced cell death in B16F10 and CT26 cells was not regulated by the ER stress response, autophagy, or JNK activation [
42,
45‐
48].
Acknowledgements
The recombinant protein ILz:rhTRAIL was purified by Ji Hye Han. BMK cells were provided by Dr. John Hiscott.