Xenografts and cell lines
The following NPC cell lines were used throughout the study: C666-1, C15, C17, C18 (EBV positive cell lines), CNE1 and HONE1 (EBV negative cell lines). A panel of non-NPC epithelial malignant and non malignant cell lines was used as control for TLR3 expression assessment: FaDu and SQ20B (head and neck squamous cell carcinoma, HNSCC), HeLa (cervical carcinoma), A431 (vulvar squamous carcinoma), A549 (non-small cell lung cancer), C33 and CaSki (cervical carcinomas), NP69 and NP460 (derived from non-malignant nasopharyngeal epithelial cells immortalized by SV-40 infection and stable transfection of the gene encoding the telomerase catalytic subunit hTert, respectively)
[20–26]. The non-malignant nasopharyngeal NP69 cell line was also used as a control for proliferation assays.
C666-1 cells are EBV-positive NPC cells which have been grown for a long time either as xenografted tumors or in vitro cultures
. Through this study, we used C666-1 cells stably transfected with the luciferase 1 gene which were kindly provided by Dr Fei-Fei Liu (university of Toronto, Ontario, Canada)
[27, 28]. These cells retain the EBV genome and intense expression of the EBER viral non-coding RNAs (see the result section). Because the luciferase gene is very stable in these cells both in vitro and in vivo, it allows in vivo imaging of the xenografted tumors. Therefore, we chose to use them from the beginning in anticipation of future in vivo studies about the effects of TLR3 agonists on NPC cells. C666-1 cells were routinely propagated in vitro using RPMI 1640 medium (Gibco-Invitrogen, Carlsbad, CA) supplemented with 25 mM HEPES and 7.5% fetal calf serum (FCS), in plastic flasks coated with collagen I (Biocoat; Becton-Dickinson, Franklin Lakes, NJ). C15, C17 and C18 are EBV-positive NPC xenografts propagated by subcutaneous passages into nude mice
. For a long time, it has not been possible to derive long-term in vitro cultures from any of these three xenografts. However, we recently adapted C17 cells to permanent in vitro propagation using a protocol inspired from Liu et al.
. Briefly, C17 xenografted tumors were minced and treated with type II collagenase for cell dispersion as previously reported
. Cells were then plated on a non-irradiated feeder layer of Normal Human Dermal Fibroblasts (NHDF; Promocell, Heidelberg, Germany) and grown in RPMI 1640 medium (Gibco-Invitrogen) supplemented with 25 mM HEPES, 7.5% fetal calf serum (FCS), and 7 μmol/L of the Rho kinases I and II inhibitor Y-27632 (Y-27632; Enzo Life Sciences, Lausen, Switzerland)
. Feeder cells became rapidly senescent. Most of them were already eliminated beyond the third in vitro passage. For cytological analysis, C17 cells were stained with hematoxilin and eosin safran (HES) after cytospin preparation. Detection of the EBERs by in situ hybridization on C666-1, HeLa, and C17 cell pellets was performed using the INFORM EBER Probe (Ref 800–2842) and the ISH iVIEW Blue Detection Kit (Ref 800–092) from Ventana-Roche (Tucson, AZ). EBV-negative cell lines CNE1 and HONE1 were grown in RPMI 1640 medium (Gibco-Invitrogen) supplemented with 5% FCS
[32, 33]. NP69 cells were grown in keratinocyte serum-free medium (Gibco) supplemented with 10% FCS.
Clinical specimens and immunohistochemistry
Biopsies were obtained from 10 patients referred to the Lariboisière hospital (Paris, France). All patients had non-keratinizing undifferentiated (or type III) NPC according to the WHO classification (2005). Biopsies were fixed in formaldehyde and paraffin-embedded. Tissue sections were microwaved at 98°C for 30 minutes in citrate buffer (10 mM, pH 7.3) and then incubated with an antihuman TLR3 mouse monoclonal antibody (40 F9.6, Innate Pharma, Marseille). Binding of the primary antibody was detected with the CSA II kit from Dako (based on a tyramide amplification system; DakoCytomation, Glostrup, Denmark). C666-1 and NP69 cell pellets embedded in paraffin were used for positive and negative control of TLR3 immunostaining. All the clinical samples were obtained and processed according to the guidelines of Lariboisière hospital institutional review board. requiring written informed consent from patients for publication.
Treatments of cells with pharmacological reagents
The polycyclic C2-symmetric (40 carbon atoms) compound RMT5265 mimics the three-dimensional structure of the N-terminal tetrapetide of Smac/Diablo (second mitochondria–derived activator of caspases)
. This compound was kindly provided by Xiaodong Wang, Dallas. It was dissolved in DMSO. The TLR3 agonists - poly(I:C) and poly(A:U) - were obtained from InvivoGen (San Diego, CA). Cisplatinum was purchased from Sigma Aldrich (St. Quentin Fallavier, France).
Cell growth and viability assays
Cell viability was determined in a short-term assay based on the reduction of MTT (CNE1, HONE1, NP69, C17) or WST (a soluble form of MTT; C666-1). MTT and WST were purchased from Sigma Aldrich. For this assay, cells were seeded in 96-well plates at a density of 2 x 103 (CNE1, HONE1, NP69) or 3 × 104 (C666-1, C17) cells per well. The MTT/WST reaction was performed after 72 hours of culture. The absorbance (Optical Density (OD)) was measured at 550 nm and 450 nm for MTT and WST assays, respectively. The percentage of inhibition was determined based on the difference of OD between treated and untreated cells, after subtraction of the optical background.
Alternatively, in vitro growth assays at low density were performed to evaluate the clonogenic potential of NPC cells. HONE1, CNE1 and NP69 cells were plated in six-well plates (5x102 cells per well) and treated at day 1 with RMT5265 and/or poly(I:C) / poly(A:U). After 2 weeks of culture, cell colonies were stained with a solution of Crystal Violet (Sigma Aldrich) in methanol. The clonal growth of C666-1 NPC cells was assessed using a feeder layer of Normal Human Dermal Fibroblasts (NHDF, Promocell). The first day of clonogenic assay, NHDF were plated in six-well plates (1.5x105 per well). 24 hours later, they were irradiated (40 Gy) and C666-1 NPC cells were added at a density of 5x103 cells per well. After 24 hours, once the epithelial cells had firmly adhered to the plate, RMT5265 and/or poly(I:C) / poly(A:U) were added to the culture medium. Replacement by fresh medium was done once a week. After 2 to 4 weeks of culture, cell colonies were stained with a solution of Rhodanile Blue (Sigma Aldrich) in ethanol. Dried plates were then scanned and digitized to allow optical magnification and precise counting of cell colonies.
We used the Bliss additivism model to quantify the level of synergy in our drug-combination experiments
. The predictive Bliss additive effect C of two single compounds with effects A and B is: C = A + B - A·B. The excess over Bliss additivism (EOBA) was calculated by subtracting the predicted Bliss additive effect from the experimentally observed inhibition. It is expressed as a percentage, and represents the excess of inhibition over the predicted response that was obtained when 2 compounds were used in combination.
Assessment of PARP cleavage
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 (see next paragraph).
Cell protein extraction and western blot analysis
Proteins from cultured cells or xenografts were extracted by lysis in RIPA buffer (50 mM Tris, 150 mM NaCl, 5 mM EDTA, 0.5% sodium deoxycholic acid, 0.5% NP-40, 0.1% SDS) supplemented with a protease inhibitor cocktail (Complete; Roche Molecular, Neuilly sur Seine, France). They were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (Immobilon, Millipore, Billerica, CA) by electroblot at 4°C for 90 minutes at 90 V or overnight at 45 V. The antibodies used for Western blot analysis were mouse monoclonal antibodies directed against the human TLR3 (clone 512505, ref. MAB1487, R&D Systems), PARP (C-2-10, ref. 53643; Santa Cruz Biotechnology), β-Actin (AC-74; Sigma Aldrich) and α-Tubulin (B-5-1-2; Sigma Aldrich). Blotted membranes were incubated with a secondary peroxidase-conjugated antibody, and chemiluminescent detection was done using the Immobilon Western Chemiluminescent HRP Substrate (Millipore, Billerica, CA). Specific protein bands were quantified using ImageQuant TL software (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) when the acquisition was performed with ImageQuant LAS 4000 mini biomolecular imager (GE Healthcare Bio-Sciences AB), and a GS-710 calibrated imaging densitometer with Quantity One software (Biorad, Marnes la Coquette, France) when detection was performed on chemiluminescence films.