Title Poly(I:C) induces intense expression of c-IAP2 and cooperates with an IAP inhibitor in induction of apoptosis in cancer cells

Background: There is increasing evidence that the toll-like receptor 3 (TLR3) is an interesting target for anti-cancer therapy. Unfortunately, most laboratory investigations about the impact of TLR3 stimulation on human malignant cells have been performed with very high concentrations 5 to 100 μg/ml of the prototype TLR3 ligand, poly(I:C). In a previous study focused on a specific type of human carcinoma nasopharyngeal carcinoma we have shown that concentrations of poly(I:C) as low as 100 ng/ml are sufficient to induce apoptosis of malignant cells when combined to a pharmacological antagonist of the IAP family based on Smac mimicry. Methods: This observation prompted us to investigate the contribution of the IAP family in cell response to poly(I:C) in a variety of human malignant cell types. Results: We report a rapid, intense and selective increase in c-IAP2 protein expression observed under stimulation by poly(I:C)(500 ng/ml) in all types of human malignant cells. In most cell types, this change in protein expression is underlain by an increase in c-IAP2 transcripts and dependent on the TLR3/TRIF pathway. When poly(I:C) is combined to the IAP inhibitor RMT 5265, a cooperative effect in apoptosis induction and/or inhibition of clonogenic growth is obtained in a large fraction of carcinoma and melanoma cell lines. Conclusions: Currently, IAP inhibitors like RMT 5265 and poly(I:C) are the subject of separate therapeutic trials. In light of our observations, combined use of both types of compounds should be considered for treatment of human malignancies including carcinomas and melanomas. Background Toll-like receptor 3, a membrane receptor of double strand RNAs, is a major effector of the immune response against viral pathogens at the cellular and systemic level. It is involved in early activation of NK and dendritic cells. It is also expressed in a wide range of non-immune cells where it plays a key role in the induction of interferon response [1]. TLR3 is frequently expressed by malignant cells of various types and there are several observations suggesting that it can be targeted for therapeutic purpose [2,3]. At least one clinical trial has shown a therapeutic benefit for breast carcinoma patients treated with the synthetic TLR3 agonist poly(A/U) [4]. On the other hand, several in vitro studies have reported apoptosis induction in malignant cells treated with the synthetic TLR3-agonist, poly(I:C). However, these results were obtained using very high concentrations of this agent in the range of 10 to 100 μg/ml [5-9]. Such concentrations are probably incompatible with doses of synthetic ligands acceptable for patient treatment. One of our previous study focused on nasopharyngeal carcinoma has opened news perspectives in this field [10]. Nasopharyngeal carcinoma or NPC is a human epithelial tumor whose malignant cells are latently infected by the Epstein-Barr virus (EBV). Using our experimental model of NPC, we could demonstrate that massive caspase-dependent apoptosis was induced in NPC cells by poly(I:C) at a low concentration (500 ng/ml) when it was combined to RMT 5265 (100 * Correspondence: [email protected] 1 Univ Paris-Sud, CNRS-UMR 8126 and Institut de Cancérologie Gustave Roussy, 39 rue Camille Desmoulins, F-94805 Villejuif, France Full list of author information is available at the end of the article © 2010 Friboulet et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Friboulet et al. BMC Cancer 2010, 10:327 http://www.biomedcentral.com/1471-2407/10/327 Page 2 of 14 nM), a synthetic inhibitor of the IAP family of proteins [10]. Inhibitor of apoptosis proteins (IAP) are a class of regulatory proteins, with mainly anti-apoptotic properties, characterized by the presence of one to three domains known as baculoviral IAP repeat (BIR) domains [11]. Among these IAP proteins, X-linked IAP (XIAP) is a direct inhibitor of caspase activity. It is produced in large amounts in all cell types and is often regarded as a housekeeping protein [11]. Cellular IAP-1 (cIAP-1) and cIAP-2 have more complex regulatory functions, many of these functions involving their E3 ubiquitin-ligase activity [1214]. Recent studies have emphasized their connection with TNF-receptor signaling and NF-kB activation [1416]. They are expressed at various levels in cancer cells depending on the tumor type [10]. Second mitochondriaderived activator of caspase (Smac) is an endogenous antagonist of IAP protein [17]. In its dimeric form, Smac, via its AVPI tetrapeptide binding motif, binds the BIR domains of XIAP, c-IAP1 and 2. It causes proteasomedependent degradation of c-IAP1 and c-IAP2 [17]. RMT 5265 is the prototype of a new class of anticancer drugs called Smac mimetics [18]. This polycyclic compound was designed for spatial mimicry of the AVPI motif of the Smac protein. It is cell permeable and specifically binds cIAP1, c-IAP2 and XIAP, triggering rapid proteasomedependent degradation of c-IAP1 and c-IAP2 [10,18]. It is also suspected to antagonize the functions of XIAP [18]. Our previous study on NPC cells provided the proof of principle 1) that synthetic TLR3 ligands could be active on malignant cells at much lower concentrations than previously reported (below 1 μg/ml); 2) that the IAP family of proteins was very important to modulate cell response to TLR3 stimulation and 3) that combinations of TLR3 ligands with IAP inhibitors were susceptible to provide a therapeutic benefit [10]. However NPC cells have unique biological features, for example a low frequency of p53 mutations, in addition to a latent EBVinfection in virtually 100% of the cases [19]. Therefore, it was important to investigate the role of IAPs in other types of malignant cells subjected to TLR3 stimulation. Using a panel of various human carcinoma and melanoma cell lines, we decided to address the following questions: 1) What is the influence of a TLR3 synthetic ligand on the status of the IAP proteins? 2) Can we enhance the pro-apoptotic effect of a TLR3 ligand by combination with an IAP inhibitor? We report that the basal concentration of c-IAP2 is at a low level in a majority of malignant cell types, in constrast with our previous observations in NPC cells. However, stimulation of TLR3 by the synthetic ligand poly(I:C) induces a rapid, substantial and specific increase in cIAP2 protein content, in all malignant cells tested. This increase is, at least to a large extent, explained by a transcriptional effect. When poly(I:C) was combined with the IAP inhibitor RMT 5265, we observed a cooperative effect in apoptosis induction and/or inhibition of clonogenic growth in a wide range of cancer cells. Methods Cell lines The following malignant cell lines were used: HeLa (epithelial, cervix carcinoma); T1, SK29 and Rosi (melanomas); SKOV3 and IGR-OV1 (ovarian carcinomas); LNCaP (prostatic carcinoma) [16,20]. MRC5 cells are non-malignant human fibroblasts purchased from Biomerieux (Marcy l'Etoile, France). The NP69 cell line was obtained by SV40 infection of epithelial cells derived from a piece of non-malignant human nasopharyngeal mucosa [21]. NP69 cells were grown in KGF keratinocyte medium supplemented with 10% fetal calf serum (FCS). LNCaP, SK29, Rosi, IGR-OV1 and MRC5 cells were continuously grown in RPMI 1640 medium (Gibco-Invitrogen, Cergy Pontoise, France) supplemented with 10% FCS, SKOV3 cells were grown in Mc Coy's 5A medium (Gibco-Invitrogen) supplemented with 10% FCS and 2% sodium bicarbonate, T1 cells in RPMI 1640 medium supplemented with 10% FCS and 1% sodium pyruvate and HeLa cells in Dulbecco's modified Eagle medium (DMEM) (Gibco-Invitrogen) supplemented with 5% FCS. Preparation of cell suspensions from tumor biopsies Fragments of tumor biopsies were obtained after written informed consent from two patients treated at the Institut de Cancérologie Gustave Roussy (Villejuif, France). Patient A, male, aged 42, had a well differentiated squamous cell carcinoma of the larynx, staged T2N0M0. Patient B, male, aged 52 had a differentiated squamous cell carcinoma of the oral cavity staged T4N2bM0. A few minutes after collection, these tumor fragments were minced in small pieces (1 to 2 mm diameter) and incubated sequentially 1) in a solution of trypsin (1.25 mg/ ml)(Sigma, St Quentin Fallavier, France) in DPBS (Dulbecco Phosphate Buffered Saline Invitrogen) at 4°C for 90 min with agitation; 2) then in a solution of collagenase (4 mg/ml)(type 2, Worthington) and DNAse (10 μg/ ml)(Sigma) in RPMI culture medium with 20% fetal calf serum at 37°C for 3 hours. Tumor cells were then dispersed by energetic pipet aspirations and releases, washed, filtered through a 70 μm cell strainer, counted and seeded in 24-well culture plates at a density of 0.5 million/well. Treatments of cells with pharmacological reagents The polycyclic C2-symmetric (40 carbon atoms) compound RMT 5265, which mimics the three-dimensionnal structure of the Smac/Diablo N-terminal tetrapeptide, has been previously described [18]. HS4044 or control Friboulet et al. BMC Cancer 2010, 10:327 http://www.biomedcentral.com/1471-2407/10/327 Page 3 of 14 agent, has a similar structure but is acetylated at a critical alanine group; it was used as a negative control [18]. Both reagents were dissolved in DMSO. Agonists of TLR3 (poly(I:C)) and TLR9 (type C CpG) were obtained from InvivoGen (Toulouse, France). Bafilomycin A1 (endosomal acidification inhibitor) and 2-Aminopurine (PKR inhibitor) were purchased from Sigma. RNA interference Expression of proteins was knocked down using the following siRNA: TRIF (TICAM1) (1: HSS152364; 2: HSS152365; 3: HSS175528), TLR3 (1: HSS110815; 2: HSS110817), c-IAP2 (HSS100561), XIAP (HSS100565). These siRNA and a negative control (NS; ref. 1390109) were purchased from Invitrogen. Transfections were carried our using Oligofectamine (Invitrogen). The final concentration of siRNAs in the culture medium was 100 nM for 4 h and 50 nM for the next 44 h. The impact on protein expression was assessed by western blotting, 48 h or 72 h after the initiation of transfection. RNA extraction, reverse transcription and polymerase reaction Total RNA was extracted from cultured cells with the RNeasy extraction kit (Qiagen, Valencia, CA). Total RNA (1 μg) was reverse-transcribed using the Protoscript First Strand cDNA Synthesis Kit (New England BioLabs, Ipswich, MA). Real-time PCR was performed in a 25 μl reaction volume, containing 25 ng of cDNA template, 10 pmol of each primer and 12.5 μl of TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA). The following sets of primers and probes were purchased from Applied Biosystems (TaqMan Gene Expression system): BIRC3 (HS00154109_m1), TLR3 (HS01551078_ m1), TLR4 (HS00152939_m1), TLR9 (HS00370913_s1), and GAPDH (HS99999905_m1). Amplification reactions were performed in an Applied Biosystems Abi Prism 7000 Sequence Detection System. Data from RQ-PCR were analysed using the comparative CT method with GADPH as an endogenous reference. Cell Growth Assays Clonogenic growth of various cell types was assessed by plating cells at low density in six well plates. Initial cell densities were established for each cell type on the basis of pilot experiments varying from 2000 (NP69) to 5000 (others cell types) per well in six well plates. After 2 to 4 weeks of culture, cell colonies were stained with a solution of crystal violet. Dried plates were then scanned and digitized to allow optical magnification and precise counting of cell colonies. Assessment of Apoptosis and Caspase Activation Apoptosis was assessed quantitatively by determining the sub-G1 DNA content in ethanol-fixed cells, stained with propidium iodide and analyzed using a Becton Dickinson FACScalibur flow cytometer and the CellQuest Pro software. Alternatively, apoptosis was evaluated by the detection of the cleavage of poly (ADP-ribose) polymerase (PARP) by western blot analysis performed on total cell protein extracts. The activities of caspases-3/7 and caspase-8 were measured with the Caspases-Glo 3/7 and Caspases-Glo 8 Assay kits, respectively (Promega, Lyon, France). These assays are based on the cleavage of luminogenic substrates containing the amino-acid sequences Z-DEVD and Z-LETD, respectively. Cell protein extraction and western blot analysis Proteins from cultured cells were extracted by lysis in RIPA buffer (50 mM Tris, 150 mM NaCl, 5 mM EDTA, 0.5% sodium DOC, 0.5% NP40, 0.1% SDS) supplemented with a protease inhibitor cocktail (Complete, Roche, Meylan, France). They were separated by SDS-PAGE and transferred to PVDF membranes (Immobilon, Millipore, Billerica, CA) by electroblotting at 4°C for 90 minutes at 90 V or overnight at 45 V. The antibodies used for western blotting were mouse monoclonal antibodies directed against human c-IAP2, XIAP and c-IAP1, obtained from BD Biosciences (reference 552782, 610763 and 556533, respectively; Le Pont de Claix, France). The other mouse monoclonal antibodies used were specific for FLICE-like inhibitory protein (FLIP) (Alexis Biochemical, ref. 804428; San Diego, CA), PARP (Santa Cruz Biotechnology; ref. 53643; Heidelberg, Germany), TRIF (Cell Signalling, ref. 4596; Danvers, MA), caspase-8 (Cell Signaling, ref. 9746), TLR3 (R&D systems ref. MAB1487; Lille, France), β-actin (Millipore, Billerica, CA, ref. MAB1501) and tubulin-α (Sigma, ref. T5168). Blots were incubated with a secondary peroxidase-conjugated antibody and chemiluminescent detection was done using the Immobilon Western Chemiluminescent HRP Substrate (Millipore, Billerica, CA). Statistics Results of real-time PCR, colony assays and caspase assays were given as the means +/standard deviation (SD). Statistical significance was assessed using a twotailed Student's t test for comparison of 2 experimental conditions. When making comparisons involving multiple experimental conditions we used the ANOVA test completed by the Tukey test (bilateral) in the XLSTAT software (confidence interval 95%). Results TLR3 transcripts are constitutively expressed in various types of human malignant cells TLR3 expression has been reported in a wide range of human malignant cells [5,6,8]. To confirm this observation, TLR3 messenger RNAs were analyzed by quantitaFriboulet et al. BMC Cancer 2010, 10:327 http://www.biomedcentral.com/1471-2407/10/327 Page 4 of 14 tive RT-PCR in a series of human malignant cell lines including one melanoma (T1), two ovarian carcinomas (IGR-OV1 and SKOV3), one endocervical carcinoma (HeLa), one prostate carcinoma (LNCaP). Two human control cell lines were analyzed in parallel: a human diploid fibroblast cell line (MRC5) and an SV40-immortalized non-tumorigenic cell line derived from human nasopharyngeal epithelial cells (NP69). As shown in Figure 1A, transcripts of TLR3 were readily detected in all 5 malignant cell lines at a higher level than in non-tumorigenic MRC5 and NP69 cells. The distribution of TLR4 and 9 transcripts was quite different. There were homogeneous high levels of TLR9 transcripts in tumorigenic and non-tumorigenic cells and a more restricted expression of TLR4 with high levels in T1 and NP69 cells. In some experimental systems, double strand RNA has been shown to induce a positive feed-back on TLR3 expression [22]. We investigated whether the same positive feedback mechanism could happen in our cell lines under treatment by poly(I:C) at 500 ng/ml for 16 h. As shown in Figure 1B, we observed almost no changes in TLR4 and TLR9 transcript levels. In contrast, TLR3 expression was enhanced more than 2 fold in 4 of the 5 malignant cell lines and only slightly modified in MRC5 and NP69. The increase in TLR3 messenger RNAs resulted in an increase in protein expression as shown in Figure 1C. In summary, TLR3 transcripts are abundant and highly inducible in various types of human malignant cells whereas they are at a low level and not inducible in two different types of human non-malignant cells. Poly(I:C) treatment increases c-IAP2 protein concentration in various types of human malignant cells In our previously published tumor model nasopharyngeal carcinoma (NPC) c-IAP2 cell concentrations were uniformly very high whereas c-IAP1 was consistently undetectable [10]. In contrast, we knew from preliminary experiments that the basal levels of c-IAP1 and c-IAP2 were highly variable in various types of non-NPC malignant cells. Therefore we decided to assess the status of several IAP proteins c-IAP1, c-IAP2 and XIAP in basal conditions and under treatment by poly(I:C). The panel of cell lines described in the previous experiment completed by two additional melanoma cell lines (SK29 and Rosi) was subjected to stimulation by poly(I:C), 500 ng/ ml for 16 h. Then the cell concentrations of IAPs were compared by western blotting in the absence or in the presence of poly(I:C). As shown in Figure 2A, in basal conditions, c-IAP2 was at a low level of concentration in most malignant cell types. However, in the presence of poly(I:C), its concentration was substantially increased in all types of malignant cells with amplitudes varying from 2 (SKOV3) to 10 fold (SK29). In contrast c-IAP2 concentrations were not modified in NP69 and MRC5 (in the latter, c-IAP2 remained undetectable). Simultaneously, the levels of c-IAP1 and XIAP concentrations were not significantly modified except a 70% increase in XIAP concentration in the SK29 cell line. The dose-effect relationships and kinetics of the increase in c-IAP2 protein expression was investigated in HeLa cells. An increase in c-IAP2 protein concentration was detectable with concentrations of poly(I:C) as low 50 ng/ml; it was dose-dependent from 50 to 250 ng/ml and then reached a plateau (Figure 2B). Stimulation by a TLR9 agonist in the same range of concentrations had absolutely no effect on c-IAP2 concentration. c-IAP1 concentrations were not modified either by poly(I:C) or the TLR9 agonist. In terms of kinetics, the increase in c-IAP2 cell concentration was detectable between 4 and 8 hours with a peak between 16 h and 24 h (Figure 2C). Overall, these data demonstrate a specific increase in c-IAP2 cell concentrations induced by treatment with poly(I:C). This change is restricted to malignant cell types. In HeLa cells it is maximal between 16 h and 24 h and detectable with concentrations of poly(I:C) as low as 50 ng/ml. To provide evidence that the stimulatory effect of poly(I:C) on cIAP2 expression was not restricted to malignant cell lines permanently propagated in vitro, we prepared tumor cells derived from Head and Neck squamous cell carcinomas and used them in short term primary cultures. These cells directly explanted from fresh tumors were treated with poly(I:C) (500 ng/ml) for 16 h. As shown in Figure 2D, we observed the same dramatic increase in c-IAP2 concentration. Poly(I:C) treatment specifically enhances c-IAP2 expression through TLR3 stimulation In order to understand how the cell concentration of cIAP2 was increased by poly(I:C), we investigated its influence on the transcription of the c-IAP2 gene (BIRC3) in 5 malignant and 2 non-malignant cell types by quantitative reverse-PCR. Under stimulation by poly(I:C) 500 ng/ml, a statistically significant increase in cell concentrations of the BIRC3 transcripts was observed in 4 of 5 malignant cell types contrasting with virtually no increase in the non-malignant cell types (Figure 3A). In terms of kinetics, HeLa cells showed a dramatic increase in the amount of BIRC3 transcripts detected as early as 2 hours after the onset of the treatment by poly(I:C) with a plateau reached at 16 hours (Figure 3B). There are several known cellular receptors or sensors for double-strand RNA including TLR3 or PKR (protein kinase R) [23]. PKR is selectively inhibited by 2-aminopurine (2-AP) whereas TLR3 is inhibited by bafilomycine A1 (BFA). BFA is an inhibitor of the endosomal acidification pathway which prevents the adequate function of TLR3 within the endosomal compartment [23]. HeLa cells were stimulated by poly(I:C) (500 ng/ml) or TNF-α (20 ng/ml) which is a Friboulet et al. BMC Cancer 2010, 10:327 http://www.biomedcentral.com/1471-2407/10/327 Page 5 of 14 Figure 1 Assessment of TLR 3, 4 and 9 messenger RNAs and TLR3 protein in a panel of human cell lines in basal conditions and under stimulation by poly(I:C). A) Relative expression levels of TLR3, 4 and 9 transcripts in basal conditions. Levels of mRNAs in HeLa cells were chosen as a reference and arbitrarily set at 1. B) TLR mRNA relative increase in expression under stimulation by poly(I:C) (500 ng/ml) for 16 h. For each cell type, the basal level of expression was taken as a reference and set at 1. C) Western blot detection of the TLR3 protein in HeLa cells treated for 16 h with increasing concentrations of poly(I:C)(from 250 ng/ml to 2 μg/ml). β-actin staining was used to control protein loading. Data presented in A), B) and C) are representative of at least two similar experiments. The stars indicate a statistical difference from respective controls (p < 0.05). A