Search

// Mathieu Veyrat 1 , Sylvere Durand 2 , Marion Classe 3 , Tanja Matijevic Glavan 4 , Natalie Oker 1, 5 , Nikiforos-Ioannis Kapetanakis 1 , Xiaojun Jiang 1 , Aurore Gelin 1 , Philippe Herman 5 , Odile Casiraghi 6 , David Zagzag 7 , David Enot 2 , Pierre Busson 1 , Benjamin Verillaud 1, 5 1 University Paris-Sud (Paris 11), CNRS-UMR 8126, Gustave Roussy, Villejuif, France 2 Equipe 11 Labelisee par la Ligue Nationale Contre le Cancer, INSERM U1138, Centre de Recherche des Cordeliers, Paris, France, Metabolomics and Molecular Cell Biology Platforms, Gustave Roussy, Villejuif, France 3 Department of Pathology, Lariboisiere Hospital, AP-HP, University Paris-Diderot Paris 7, Paris, France 4 Division of Molecular Medicine, Rudjer Boskovic Institute, Zagreb, Croatia 5 Department of Head and Neck surgery, Lariboisiere Hospital, AP-HP, University Paris-Diderot Paris 7, Paris, France 6 Department of Biopathology, Gustave Roussy, Villejuif, France 7 Department of Neuropathology, New York University School of Medicine, New York, NY, USA Correspondence to: Benjamin Verillaud, email: benjamin.verillaud@gmail.com Keywords: toll-like receptor 3, innate immunity, warburg effect, HIF, metabolomics Received: February 12, 2015 Accepted: October 19, 2016 Published: October 25, 2016 ABSTRACT In this study, a possible link between the innate immune recognition receptor TLR3 and metabolic reprogramming in Head and Neck carcinoma (HNC) cells was investigated. The effects of TLR3 stimulation/knock-down were assessed under several culture conditions in 4 HNC cell-lines by cell growth assays, targeted metabolomics, and glycolysis assays based on time-resolved analysis of proton release (Seahorse analyzer). The stimulation of TLR3 by its synthetic agonist Poly(A:U) resulted in a faster growth of HNC cells under low foetal calf serum conditions. Targeted analysis of glucose metabolism pathways demonstrated a tendency towards a shift from tricarboxylic acid cycle (Krebs cycle) to glycolysis and anabolic reactions in cells treated with Poly(A:U). Glycolysis assays confirmed that TLR3 stimulation enhanced the capacity of malignant cells to switch from oxidative phosphorylation to extra-mitochondrial glycolysis. We found evidence that HIF-1α is involved in this process: addition of the TLR3 agonist resulted in a higher cell concentration of the HIF-1α protein, even in normoxia, whereas knocking-down TLR3 resulted in a lower concentration, even in hypoxia. Finally, we assessed TLR3 expression by immunohistochemistry in a series of 7 HNSCC specimens and found that TLR3 was detected at higher levels in tumors displaying a hypoxic staining pattern. Overall, our results demonstrate that TLR3 stimulation induces the Warburg effect in HNC cells in vitro , and suggest that TLR3 may play a role in tumor adaptation to hypoxia.

Background Because latent Epstein Barr (EBV)-infection is a specific characteristic of malignant nasopharyngeal carcinoma (NPC), various molecules of viral origin are obvious candidate biomarkers in this disease. In a previous study, we could show in a few clinical samples that it was possible to detect a category of EBV microRNAs called miR-BARTs in the plasma of at least a fraction of NPC patients. The first aim of the present study was to investigate the status of circulating miR-BART17-5p (one of the miR-BARTs hereafter called miR-BART17) and EBV DNA in a larger series of NPC plasma samples. The second aim was to determine whether or not circulating miR-BART17 was carried by plasma exosomes. Patients and methods Plasma samples were collected from 26 NPC patients and 10 control donors, including 9 patients with non-NPC Head and Neck squamous cell carcinoma and one healthy EBV carrier. Concentrations of miR-BART17 and two cellular microRNAs (hsa-miR-16 and -146a) were assessed by real-time quantitative PCR with spike-in normalization and absolute quantification. In addition, for 2 patients, exosome distributions of miR-BART17 and miR-16 were investigated following plasma lipoprotein fractionation by isopycnic density gradient ultrcentrifugation. Results The miR-BART17 was significantly more abundant in plasma samples from NPC patients compared to non-NPC donors. Above a threshold of 506 copies/mL, detection of miR-BART17 was highly specific for NPC patients (ROC curve analysis: AUC=0.87 with true positive rate = 0.77, false positive rate = 0.10). In this relatively small series, the concentration of plasma miR-BART17 and the plasma EBV DNA load were not correlated. When plasma samples were fractionated, miR-BART17 co-purified with a protein-rich fraction but not with exosomes. Conclusions Detection of high concentrations of plasma miR-BART17 is consistent in NPC patients. This parameter is, at least in part, independent of the viral DNA load. Circulating miR-BART17 does not co-purify with exosomes.

Background Nasopharyngeal carcinoma (NPC) is a human epithelial malignancy consistently associated with the Epstein-Barr virus. The viral genome is contained in the nuclei of all malignant cells with abundant transcription of a family of viral microRNAs called BART miRNAs. MicroRNAs are well known intra-cellular regulatory elements of gene expression. In addition, they are often exported in the extra-cellular space and sometimes transferred in recipient cells distinct from the producer cells. Extra-cellular transport of the microRNAs is facilitated by various processes including association with protective proteins and packaging in secreted nanovesicles called exosomes. Presence of microRNAS produced by malignant cells has been reported in the blood and saliva of tumor-bearing patients, especially patients diagnosed with glioblastoma or ovarian carcinoma. In this context, it was decided to investigate extra-cellular release of BART miRNAs by NPC cells and their possible detection in the blood of NPC patients. To address this question, we investigated by quantitative RT-PCR the status of 5 microRNAs from the BART family in exosomes released by NPC cells in vitro as well as in plasma samples from NPC xenografted nude mice and NPC patients. Results We report that the BART miRNAs are released in the extra-cellular space by NPC cells being associated, at least to a large extent, with secreted exosomes. They are detected with a good selectivity in plasma samples from NPC xenografted nude mice as well as NPC patients. Conclusions Viral BART miRNAs are secreted by NPC cells in vitro and in vivo. They have enough stability to diffuse from the tumor site to the peripheral blood. This study provides a basis to explore their potential as a source of novel tumor biomarkers and their possible role in communications between malignant and non-malignant cells.

6079 Background: Epstein-Barr virus (EBV) microRNAs of the BART family can be detected in the plasma of at least a fraction of nasopharyngeal carcinoma (NPC) patients. Our aim was to investigate the specificity of plasma BART detection in NPC patients and the correlation with viral DNA load and/or clinical characteristics. Methods: Hsa miR484 and 4 miR-BARTs (ebv-miR BART 2-5p, 5, 9 and 18) were assessed by RT-PCR following RNA extraction in 37 plasma samples including 20 NPC, 8 non-NPC head and neck squamous cell carcinoma (SCC) patients and 9 healthy EBV carriers. EBV DNA copy numbers (viral load) were measured in the same samples. To explore correlations of BART concentrations with tumor mass, MRI-based volume analysis was designed and performed on a subset of 13 patients with non metastatic NPC. In addition, longitudinal variations of BART concentrations were studied in 9 NPC patients for whom we had sequential plasma samples. Results: The cellular hsa miR-484 was detected in all plasma samples with a slightly higher average concentration in plasma samples from NPC and SCC patients. On the other hand, the miR-BARTs were undetectable or at a very low level in samples from SCC patients and healthy carriers contrasting with a high level in most NPC patients. There was an overall positive correlation between EBVDNA copy numbers and plasma concentrations of each miR-BART. However substantial amounts of plasmatic miR-BARTs were observed in several patients for whom no plasma EBV DNA was detected.Their concentration was apparently not correlated to the initial tumor volume but their longitudinal variations were parallel with clinical evolution in all but one case. Conclusions: To a large extent, detection of plasma miR-BART-2-5p, 5, 9, and 18 is specific of NPC patients. Their concentration does not seem to correlate with tumor mass, at least in localized NPC patients. In a fraction of patients without initial detection of plasma EBV DNA, the kinetics of plasma BARTs may provide informations on early tumor response under treatment and may have prognostic value.