Therapeutic inhibition of TRF1 impairs the growth of p53-deficient K-RasG12V-induced lung cancer by induction of telomeric DNA damage

Telomeres are considered anti-cancer targets, as telomere maintenance above a minimum length is necessary for cancer growth. Telomerase abrogation in cancer-prone mouse models, however, only decreased tumor growth after several mouse generations when telomeres reach a critically short length, and this effect was lost upon p53 mutation. Here, we address whether induction of telomere uncapping by inhibition of the TRF1 shelterin protein can effectively block cancer growth independently of telomere length. We show that genetic Trf1 ablation impairs the growth of p53-null K-RasG12V-induced lung carcinomas and increases mouse survival independently of telomere length. This is accompanied by induction of telomeric DNA damage, apoptosis, decreased proliferation, and G2 arrest. Long-term whole-body Trf1 deletion in adult mice did not impact on mouse survival and viability, although some mice showed a moderately decreased cellularity in bone marrow and blood. Importantly, inhibition of TRF1 binding to telomeres by small molecules blocks the growth of already established lung carcinomas without affecting mouse survival or tissue function. Thus, induction of acute telomere uncapping emerges as a potential new therapeutic target for lung cancer.

plated and infected 3 times with pBabeCre viral supernatant supplemented with 4 µg/mL polybrene. MEFs were then selected with 2 µg/mL puromycin for 3 days. For all the assays performed, they were kept on DMEM with puromycin supplemented with 2% fetal calf serum. Trf1 excision was monitored by PCR as previously described (Martinez et al, 2009). For proliferation assays, 5 X 10 4 cells were plated on six-well plates, with duplicates. Their growth rate was determined using an automatic cell counter (Millipore Scepter Automatic Cell Counter), on days 2, 3, 4, 5 and 7 postplating. For colony formation assays, 5000 cells were seeded on 10-cm plates, with duplicates. After 2 weeks, cells were fixed and stained with GIEMSA. Colonies were counted and measured. !-galactosidase senescence-associated activity on day 7 was detected using a commercial kit (Cell Signalling). For proliferation assay CellTiter 96®AQueousOne Solution Cell Proliferation Assay (G3582, Promega) was used according to the manufactures protocol For allograft and xenograft experiments, K-Ras !/LG12Vgeo p53 -/tumor-derived cell line (M. Barbacid's gift) and the A549 human carcinoma cell line (ATCC nº; were infected with a shRNA against Trf1 (pLKO.1-puro-Trf1 shRNA, Sigma-Aldrich) as described above. Trf1 expression was quantified by qPCR using the following primers: TRF1 FB (5' TCT AAG GAT AGG CCA GAT GCC A 3') and TRF1 RB (5'CTG AAA TCT GAT GGA GCA CGT C 3'. Actin was used as a house keeping gene. We determined the relative expression of Trf1 in each sample by calculating the 2!CT value. For each sample, 2!CT was normalized to the control 2!CT mean.

Western blotting
Western blot was performed on MEFs pellets from day 7 following standard procedures. The antibody used was raised against active Caspase 3 (Asp 175, Cell Signaling) and anti-Trf1 (ab10579, Abcam). For iPS cells, 20"g of nuclear extracts of untreated control (DMSO) or treated  were resolved in 4-12% SDS/PAGE gels (NuPAGE Invitrogen) and transferred to nitrocellulose membranes. Blots were incubated with the primary antibodies anti-TRF1 (raised at the Monoclonal Antibodies Unit at CNIO, against full-length mouse TRF1 protein) or anti-SMC1 (Bethyl Laboratories #A300-055A) used as loading control.

Histopathology, Immunohistochemistry and immunofluorescence techniques
For quantification and classification of tumor lesions, 4/5 of lung lobes were fixed in 10% buffered formalin (Sigma) and embedded in paraffin. The remaining 1/5 of lung lobe was processed for whole-mount X-Gal staining to detect !-Geo, as a surrogate marker for K-Ras G12V expression (Guerra et al, 2003). Stained tissues were embedded in paraffin, serially sectioned, and tumors counted and analyzed by a pathologist. Following the classical histopathology criteria for the diagnosis of malignancy (local/vascular invasion and metastasis), the pathologist classified the tumors into adenomas (benign) and adenocarcinomas (malignant). Due to the high cellular atypia (anaplasia) observed in the Trf1 !/! K-Ras +/G12V p53 -/tumors, pure lepidic growth lacking invasion, regardless of the cellular atypia was classified as benign.
Three sections of each lung were digitally scanned (Mirax Panoramic Scanner, 3DHistech) for lesion size measure with Panoramic Viewer software.

Pharmacokinetic study of ETP-47037
Female BABL/c mice, 10 weeks old were used (n=3 per time point) for the pharmacokinetic studies of ETP-47037. The compound was formulated in 10 %Nmethyl-pyrrolidone and 90 % polyethylene-glycol 300 for oral administrations and in 10 %N-methyl-pyrrolidone, 50 % polyethyleneglycol 300 and 40 % glucose at 5 % in water for intravenous (i.v.) injections. Plasma samples were collected following a single i.v. or oral administration of 3 mg/kg at 0.08, 0.25, 0.5, 1, 4, 8 hours. The analysis of ETP-47037 from plasma was achieved by solid phase extraction followed by high performance liquid chromatography/ tandem mass spectrometry (Agilent 1100, Applied Biosystems API2000) analysis. The amount of inhibitor and the internal standard in each mouse plasma sample were quantified based on calibration curves generated using standards of known concentrations of compound. This assay method was sufficiently accurate for the quantification of ETP-47037 with a limit of detection (LOD) and a low limit of quantification (LLOQ) of 5 ng/mL. Pharmacokinetic parameters were estimated using WinNonlin Version 5.2 software (Pharsight Corp., CA), by fitting both the experimental i.v and oral data to a bicompartimental model. foci. Percentage of low vs. High foci intensity was calculated. Compounds significantly decreasing percentage of high intensity foci were considered as positives hits and were taken for further validation. (C) The Z'-factor coefficient, a statistical parameter that in addition to consider the window in the assay also considers the variance around both the high and low signals in the assay , it is commonly used to assess the robustness of high throughput screening (HTS) assays. The z' factor was calculated as follow: Z'-factor=1-3X(sp+sn)/|mp-mn|,(1) m: mean fluorescence intensity and s: standard deviation; n:negative control (sh-Trf1 or eGFP-Trf1+/ki heterozygous, minimum signal) and p:positive (homozygous eGFP-Trf1, maximum signal). Z'-Factors of (Homo vs.