Search

Claire Fonti and Anne Saumet contributed equally to this work.; International audience; Molecular subtypes of breast cancer are defined on the basis of gene expression and genomic/epigenetic pattern differences. Different subtypes are thought to originate from distinct cell lineages, but the early activation of an oncogene could also play a role. It is difficult to discriminate the respective inputs of oncogene activation or cell type of origin. In this work, we wished to determine whether activation of distinct oncogenic pathways in human mammary epithelial cells (HMEC) could lead to different patterns of genetic and epigenetic changes. To this aim, we transduced shp53 immortalized HMECs in parallel with the CCNE1, WNT1 and RASv12 oncogenes which activate distinct oncogenic pathways and characterized them at sequential stages of transformation for changes in their genetic and epigenetic profiles. We show that initial activation of CCNE1, WNT1 and RASv12, in shp53 HMECs results in different and reproducible changes in mRNA and micro-RNA expression, copy number alterations (CNA) and DNA methylation profiles. Noticeably, HMECs transformed by RAS bore very specific profiles of CNAs and DNA methylation, clearly distinct from those shown by CCNE1 and WNT1 transformed HMECs. Genes impacted by CNAs and CpG methylation in the RAS and the CCNE1/WNT1 clusters showed clear differences, illustrating the activation of distinct pathways. Our data show that early activation of distinct oncogenic pathways leads to active adaptive events resulting in specific sets of CNAs and DNA methylation changes. We, thus, propose that activation of different oncogenes could have a role in reshaping the genetic landscape of breast cancer subtypes.

Gene expression differences, combined with distinct patterns of genomic rearrangements and epigenetic modifications, have laid the bases of molecular classification of breast cancer. Different molecular subtypes are thought to originate from different cell lineages in the mammary gland, but the early activation of an oncogene could also play a role. It is, however, difficult to discriminate the respective inputs of oncogene activation or cell type of origin in the natural history of the tumor. In this work, we have designed an experimental strategy aiming at determining whether activation of distinct oncogenic pathways in human mammary epithelial cells (HMEC) could lead to different patterns of genetic and epigenetic changes. We show that initial activation of CCNE1, WNT1 and RASv12, which activate distinct oncogenic pathways, in shp53 immortalized HMECs results in different and reproducible profiles of mRNA and miRNA expression, copy number alterations (CNA) and DNA methylation modifications. Noticeably, HMECs transformed by RAS bore very specific profiles of CNAs and DNA methylation, clearly distinct from those shown by CCNE1 and WNT1 transformed HMECs.Genes impacted by CNAs and CpG methylation in the RAS and the CCNE1/WNT1 clusters showed clear differences, illustrating the activation of distinct pathways. Our data show that early activation of distinct oncogenic pathways leads to active adaptive events resulting in specific sets of CNAs and DNA methylation changes. We, thus, propose that activation of different oncogenes could have a role in reshaping the genetic landscape of breast cancer subtypes.Author summaryGenetic and epigenetic changes are at the center of cancer development. Breast cancer molecular subtypes are defined on differences in genetic and epigenetic profiles and it is generally assumed these subtypes originate from different cell lineages in the mammary gland. We propose that founding oncogenic mutations could also have an impact. To address this question, we designed an experimental model, based on the ectopic expression of different oncogenes in human mammary epithelial cells (HMEC), and monitored genetic and DNA methylation changes occurring at different stages of cell transformation. We show that transformation of HMEC by distinct oncogenes resulted in clearly different and reproducible patterns of genetic and DNA methylation changes. Genes whose expression was modified by either CNAs or CpG methylation were consistent with the dominant pathways activated and reflected the phenotypes in the respective models. We propose that DNA methylation and CNA changes correspond to adaptive responses to the activation of the oncogenic pathways. Our data strongly suggest that early activation of distinct oncogenic insults will not only impinge on the phenotypic characteristics of the resulting tumors, but also have a strong impact on their genomic and epigenetic landscapes.

microRNA deregulations are often, if not invariably, associated with human malignancies, including cancers. Though most of these deregulations may not be functionally implicated in tumorigenesis, the fact that microRNA expression can be monitored in a variety of human specimens, including biological fluids, supports studies aimed at characterizing microRNA signatures able to detect various cancers (diagnosis), predict their outcome (prognosis), monitor their treatment (theranosis), and adapt therapy to a patient (precision medicine). Here, we review and discuss pros and cons of microRNA-based approaches that can support their exploitation as cancer biomarkers.

Gene expression differences, combined with distinct patterns of genomic rearrangements and epigenetic modifications constitute the bases of molecular classification of breast cancer. Molecular subtypes may originate from different cell lineages in the mammary gland, but also from the early activation of oncogenes that may drive the establishment of these molecular subtypes. However, in the natural history of human cancer, it is difficult to discriminate between these two factors : cell lineage and initial oncogenic alterations. In this work, we designed an experimental strategy aiming at determining whether activation of distinct oncogenic pathways in human mammary epithelial cells (HMEC) could lead to different patterns of genetic and epigenetic changes. We show that initial activation of CCNE1, WNT1 and RASv12 in shp53 immortalized HMECs results in different and reproducible profiles of mRNA and miR expression, copy number alterations (CNA) and DNA methylation modifications. Interestingly, first the extend of CNA measured as fraction of the genome altered was lower in HMECs transformed by RAS than CCNE1 and WNT1 transformed HMECs revealing a lower genetic instability. This was confirmed by less numerous γH2Ax and 53BP1 nuclear foci in HMECs transformed by RAS. Second, HMECs transformed by RAS bore specific profiles of CNAs and DNA methylation, clearly distinct of those shown by CCNE1 and WNT1 transformed HMECs. Genes differentially expressed in the RAS and the CCNE1/WNT1 clusters and included in CNAs or with variable CpG methylation were mostly different, depicting the activation of distinct signaling pathways in these two clusters. These data indicate that early activation of distinct oncogenic pathways may leads to adaptive events resulting in specific pattern of CNAs and DNA methylation changes. We postulated that the early activation of oncogenes may be an important determinant of the establishment of breast cancer molecular subtypes along with cell lineage origin. Citation Format: Stan P. du manoir, Claire Fonti, Anne Saumet, Amanda Abi-Khalil, Béatrice Orsetti, Cleroux Elouan, Ambre Bender, Michael Dumas, Jacques Colinge, Michael Weber, Charles Theillet. Distinct oncogenic events induce different DNA methylation and copy number changes in human mammary epithelial cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3112.

4; International audience; Epithelial to mesenchymal transition (EMT) is a key step toward metastasis. MCF7 breast cancer cells conditionally expressing the EMT master regulator SNAI1 were used to identify early expressed microRNAs (miRNAs) and their targets that may contribute to the EMT process. Potential targets of miRNAs were identified by matching lists of in silico predicted targets and of inversely expressed mRNAs. MiRNAs were ranked based on the number of predicted hits, highlighting miR-661, a miRNA with so far no reported role in EMT. MiR-661 was found required for efficient invasion of breast cancer cells by destabilizing two of its predicted mRNA targets, the cell-cell adhesion protein Nectin-1 and the lipid transferase StarD10, resulting, in turn, in the downregulation of epithelial markers. Reexpression of Nectin-1 or StarD10 lacking the 3'-untranslated region counteracted SNAI1-induced invasion. Importantly, analysis of public transcriptomic data from a cohort of 295 well-characterized breast tumor specimen revealed that expression of StarD10 is highly associated with markers of luminal subtypes whereas its loss negatively correlated with the EMT-related, basal-like subtype. Collectively, our non-a priori approach revealed a nonpredicted link between SNAI1-triggered EMT and the down-regulation of Nectin-1 and StarD10 through the up-regulation of miR-661, which may contribute to the invasion of breast cancer cells and poor disease outcome.

Micro(mi)RNAs are small noncoding RNAs that orchestrate many key aspects of cell physiology and their deregulation is often linked to distinct diseases including cancer. Here, we studied the contribution of miRNAs in a well-characterized human myeloid leukemia, acute promyelocytic leukemia (APL), targeted by retinoic acid and trioxide arsenic therapy. We identified several miRNAs transcriptionally repressed by the APL-associated PML-RAR oncogene which are released after treatment with all-trans retinoic acid. These coregulated miRNAs were found to control, in a coordinated manner, crucial pathways linked to leukemogenesis, such as HOX proteins and cell adhesion molecules whose expressions are thereby repressed by the chemotherapy. Thus, APL appears linked to transcriptional perturbation of miRNA genes, and clinical protocols able to successfully eradicate cancer cells may do so by restoring miRNA expression. The identification of abnormal miRNA biogenesis in cancer may therefore provide novel biomarkers and therapeutic targets in myeloid leukemias.

Many aspects of physiology and behavior in organisms from bacteria to man are subjected to circadian regulation. Indeed, the major function of the circadian clock consists in the adaptation of physiology to daily environmental change and the accompanying stresses such as exposition to UV-light and food-contained toxic compounds. In this way, most aspects of xenobiotic detoxification are subjected to circadian regulation. These phenomena are now considered as the molecular basis for the time-dependence of drug toxicities and efficacy. However, there is now evidences that these toxic compounds can, in turn, regulate circadian gene expression and thus influence circadian rhythms. As food seems to be the major regulator of peripheral clock, the possibility that food-contained toxic compounds participate in the entrainment of the clock will be discussed. (c) 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

International audience; microRNA deregulations are often, if not invariably, associated with human malignancies, including cancers. Though most of these deregulations may not be functionally implicated in tumorigenesis, the fact that microRNA expression can be monitored in a variety of human specimens, including biological fluids, supports studies aimed at characterizing microRNA signatures able to detect various cancers (diagnosis), predict their outcome (prognosis), monitor their treatment (theranosis), and adapt therapy to a patient (precision medicine). Here, we review and discuss pros and cons of microRNA-based approaches that can support their exploitation as cancer biomarkers.

By means of its antiangiogenic activity, thrombospondin-1 (TSP-1) exerts indirect antitumoral action on solid tumors. Here, we investigated potential antitumor action in an in vitro cell model for promyelocytic leukemia (NB4-LR1), resistant to retinoid maturation. Purified soluble TSP-1 added to cultures induced a strong dose-dependent growth inhibition and a slowly developing maturation-independent cell death. Recombinant fragments of TSP-1 allowed mapping of these activities to its type 3 repeat/C-terminal domain, features that are distinct from those of TSP-1 action on solid tumors, previously ascribed to the type 1 repeat domain. Cell death in leukemia was characterized as a caspase-independent mechanism, without DNA fragmentation, but phosphatidylserine externalization followed by membrane permeabilization. Mitochondria membrane depolarization was inherent to TSP-1 action but did not produce release of death-promoting proteins (eg, noncaspase apoptosis regulators, apoptosis-induced factor [AIF], endonuclease G, or Omi/HtrA2 or the caspase regulators, cytochrome c or second mitochondrial activator of caspase/direct inhibitor of apoptosis protein-binding protein with low isoelectric point [Smac/DIABLO]). Although detected, reactive oxygen species (ROS) production was likely not involved in the death process. Finally, receptor agonist RFYVVM and RGD peptides indicated that TSP-1 death effects are mediated by membrane receptors CD47 and αvβ3. These results demonstrated a new domain-specific antitumoral activity of TSP-1 on a leukemia cell line, which extends TSP-1 therapeutic potential outside the area of vascularized solid tumors. (Blood. 2005;106:658-667)

Thrombospondin-1 (TSP-1) is an adhesive glycoprotein which, when secreted from α-granules of activated platelets, can bind to the cell surface and participate in platelet aggregate formation. In this study, we show that thrombin activation leads to the rapid and specific association of a large amount of secreted α-granular TSP-1 with the actin cytoskeleton. This cytoskeletal association of TSP-1 was correlated with platelet secretion, but not aggregation, and was inhibited by cytochalasin D, an inhibitor of actin polymerization. Association of TSP-1 with the actin cytoskeleton was mediated by membrane receptors, as shown by using MAII, a TSP-1-specific monoclonal antibody that inhibited both TSP-1 surface binding to activated platelets and cytoskeletal association. TSP-1 and its potential membrane receptors, e.g. αIIbβ3 integrin, CD36 and CD47, concomitantly associated with the actin cytoskeleton. However, studies on platelets from a patient with type I Glanzmann's thrombasthenia lacking αIIbβ3 and another with barely detectable CD36 showed normal TSP-1 surface expression and association with the actin cytoskeleton. Likewise, no involvement of CD47 in TSP-1 association with the actin cytoskeleton could be inferred from experiments with control platelets using the function-blocking anti-CD47 antibody B6H12. Finally, assembly of signalling complexes, as observed through translocation of tyrosine-phosphorylated proteins and kinases to the actin cytoskeleton, was found to occur in concert with cytoskeletal association of TSP-1, in control platelets as well as in thrombasthenic and CD36-deficient platelets. Our results imply a role for the actin cytoskeleton in the membrane-surface expression process of TSP-1 molecules and suggest a possible coupling of TSP-1 receptors to signalling events occurring independently of αIIbβ3 or CD36. These results provide new insights into the link between surface-bound TSP-1 and the contractile actin microfilament system which may promote platelet aggregate cohesion.

International audience; NK cell is an innate immune system lymphocyte lineage with natural cytotoxicity. Its optimal use in the clinic requires in vitro expansion and activation. Cytokines and encounter with target cells activate NK cells and induce proliferation, and this could depend on the presence of other immune cells. Here we activated PBMCs during 5 days with IL-2, with IL-2 plus the tumor cell line K562 and with the lymphoblastoid cell line R69 and perform integrated analyses of microRNA and mRNA expression profiles of purified NK cells. The samples cluster depending on the stimuli and not on the donor, indicating that the pattern of NK cell stimulation is acutely well conserved between individuals. Regulation of mRNA expression is tighter than that of miRNA expression. All stimuli induce a common preserved genetic remodeling. In addition, encounter with target cells mainly activates pathways related to metabolism. Different target cells induce different NK cell remodeling which affects cytokine response and cytotoxicity, supporting the notion that encounter with different target cells significantly changing the activation pattern. We validate our analysis by showing that activation down regulates miR-23a, which is a negative regulator of cathepsin C (CTSC) mRNA, a gene up regulated by all stimuli. The peptidase CTSC activates the granzymes, the main effector proteases involved in NK cell cytotoxicity. All-trans retinoic acid (ATRA), which induces miR-23a expression, decreases CTSC expression and granzyme B activity leading to impaired NK cell cytotoxicity in an in vivo mouse model.

International audience; MicroRNAs orchestrate the expression of the genome and impact many, if not all, cellular processes. Their deregulation is thus often causative of human malignancies, including cancers. Numerous studies have implicated microRNAs in the different steps of tumorigenesis including initiation, progression, metastasis, and resistance to chemo/radiotherapies. Thus, microRNAs constitute appealing targets for novel anticancer therapeutic strategies aimed at restoring their expression or function. As microRNAs are present in a variety of human cancer types, microRNA profiles can be used as tumor-specific signatures to detect various cancers (diagnosis), to predict their outcome (prognosis), and to monitor their treatment (theranosis). In this review, we present the different aspects of microRNA biology that make them remarkable molecules in the emerging field of personalized medicine against cancers and provide several examples of their industrial exploitation.

RNA interference (RNAi) is a potent cellular system against viruses in various organisms. Although common traits are observed in plants, insects, and nematodes, the situation observed in mammals appears more complex. In mammalian somatic cells, RNAi is implicated in endonucleolytic cleavage mediated by artificially delivered small interfering RNAs (siRNAs) as well as in translation repression mediated by microRNAs (miRNAs). Because siRNAs and miRNAs recognize viral mRNAs, RNAi inherently limits virus production and participates in antiviral defense. However, several observations made in the cases of hepatitis C virus and retroviruses (including the human immunodeficiency virus and the primate foamy virus) bring evidence that this relationship is much more complex and that certain components of the RNAi effector complex [called the RNA-induced silencing complex (RISC)], such as AGO2, are also required for viral replication. Here, we summarize recent discoveries that have revealed this dual implication in virus biology. We further discuss their potential implications for the functions of RNAi-related proteins, with special emphasis on retrotransposition and genome stability.

International audience; In addition to estrogen receptor modulators, retinoic acid and other retinoids are promising agents to prevent breast cancer. Retinoic acid and estrogen exert antagonistic regulations on the transcription of coding genes and we evaluated here whether these two compounds have similar effects on microRNAs. Using an integrative approach based on several bioinformatics resources together with experimental validations, we indeed found that retinoic acid positively regulates miR-210 and miR-23a/24-2 expressions and is counteracted by estrogen. Conversely, estrogen increased miR-17/92 and miR-424/450b expressions and is inhibited by retinoic acid. In silico functional enrichment further revealed that this combination of transcriptional/post-transcriptional regulations fully impacts on the molecular effects of estrogen and retinoic acid. Besides, we unveiled a novel effect of retinoic acid on aerobic glycolysis. We specifically showed that it increases extracellular lactate production, an effect counteracted by the miR-210 and the miR-23a/24-2, which simultaneously target lactate dehydrogenase A and B mRNAs. Together our results provide a new framework to better understand the estrogen/retinoic acid antagonism in breast cancer cells.

International audience; Cancer cells have elevated aerobic glycolysis that is termed the Warburg effect. But several tumor cells, including leukemic cells, also increase glutamine metabolism, which is initiated by glutaminase (GLS). The microRNA (miRNA) miR-23 targets GLS mRNA and inhibits expression of GLS protein. Here we show that in human leukemic Jurkat cells the NF-κB p65 subunit binds to miR-23a promoter and inhibits miR-23a expression. Histone deacetylase (HDAC) inhibitors release p65-induced inhibition. Jurkat cells growing in glutamine decrease proliferation due to cell accumulation in G0/G1 phase. Nevertheless, cells get used to this new source of energy by increasing GLS expression, which correlates with an increase in p65 expression and its translocation to the nucleus, leading to a higher basal NF-κB activity. Jurkat cells overexpressing p65 show increase basal GLS expression and proliferate faster than control cells in glutamine medium. Overexpressing miR-23a in leukemic cells impaired glutamine use and induces mitochondrial dysfunction leading to cell death. Therefore, p65 activation decreases miR-23a expression, which facilitates glutamine consumption allowing leukemic cells to use this alternative source of carbon and favoring their adaptation to the metabolic environment.

International audience; Cellular micro(mi)RNAs are able to recognize viral RNAs through imperfect micro-homologies. Similar to the miRNA-mediated repression of cellular translation, this recognition is thought to tether the RNAi machinery, in particular Argonaute 2 (AGO2) on viral messengers and eventually to modulate virus replication. Here, we unveil another pathway by which AGO2 can interact with retroviral mRNAs. We show that AGO2 interacts with the retroviral Group Specific Antigen (GAG) core proteins and preferentially binds unspliced RNAs through the RNA packaging sequences without affecting RNA stability or eliciting translation repression. Using RNAi experiments, we provide evidences that these interactions, observed with both the human immunodeficiency virus 1 (HIV-1) and the primate foamy virus 1 (PFV-1), are required for retroviral replication. Taken together, our results place AGO2 at the core of the retroviral life cycle and reveal original AGO2 functions that are not related to miRNAs and translation repression.

Background To understand biological processes and diseases, it is crucial to unravel the concerted interplay of transcription factors (TFs), microRNAs (miRNAs) and their targets within regulatory networks and fundamental sub-networks. An integrative computational resource generating a comprehensive view of these regulatory molecular interactions at a genome-wide scale would be of great interest to biologists, but is not available to date. Results To identify and analyze molecular interaction networks, we developed MIR@NT@N, an integrative approach based on a meta-regulation network model and a large-scale database. MIR@NT@N uses a graph-based approach to predict novel molecular actors across multiple regulatory processes (i.e. TFs acting on protein-coding or miRNA genes, or miRNAs acting on messenger RNAs). Exploiting these predictions, the user can generate networks and further analyze them to identify sub-networks, including motifs such as feedback and feedforward loops (FBL and FFL). In addition, networks can be built from lists of molecular actors with an a priori role in a given biological process to predict novel and unanticipated interactions. Analyses can be contextualized and filtered by integrating additional information such as microarray expression data. All results, including generated graphs, can be visualized, saved and exported into various formats. MIR@NT@N performances have been evaluated using published data and then applied to the regulatory program underlying epithelium to mesenchyme transition (EMT), an evolutionary-conserved process which is implicated in embryonic development and disease. Conclusions MIR@NT@N is an effective computational approach to identify novel molecular regulations and to predict gene regulatory networks and sub-networks including conserved motifs within a given biological context. Taking advantage of the M@IA environment, MIR@NT@N is a user-friendly web resource freely available at http://mironton.uni.lu which will be updated on a regular basis.

The anti-viral function of RNA silencing was first discovered in plants as a natural manifestation of the artificial 'co-suppression', which refers to the extinction of endogenous gene induced by homologous transgene. Because silencing components are conserved among most, if not all, eukaryotes, the question rapidly arose as to determine whether this process fulfils anti-viral functions in animals, such as insects and mammals. It appears that, whereas the anti-viral process seems to be similarly conserved from plants to insects, even in worms, RNA silencing does influence the replication of mammalian viruses but in a particular mode: micro(mi)RNAs, endogenous small RNAs naturally implicated in translational control, rather than virus-derived small interfering (si)RNAs like in other organisms, are involved. In fact, these recent studies even suggest that RNA silencing may be beneficial for viral replication. Accordingly, several large DNA mammalian viruses have been shown to encode their own miRNAs. Here, we summarize the seminal studies that have implicated RNA silencing in viral infection and compare the different eukaryotic responses.

Human telomerase has been implicated in cell immortalization and cancer. Recent works suggest that telomerase confers additional function required for tumorigenesis that does not depend on its ability to maintain telomeres. This new action may influence tumor therapy outcomes by yet unraveled mechanisms. Here, we show that overexpression of the catalytic subunit of telomerase (hTERT) protects a maturation-resistant acute promyelocytic leukemia (APL) cell line from apoptosis induced by the tumor necrosis factor (TNF) or TNF-related apoptosis-inducing ligand (TRAIL) and not from apoptosis induced by chemotherapeutic drugs such as etoposide or cisplatin. Conversely, in these cells, TRAIL-induced cell death is magnified by all-trans retinoic acid (ATRA) treatment, independently of telomerase activity on telomeres. Of note, this response is subordinated neither to maturation nor to telomere shortening. This work underlines that retinoids and death receptor signaling cross-talks offer new perspectives for antitumor therapy.

Thrombospondin-1 (TSP-1) is an adhesive glycoprotein which, when secreted from alpha-granules of activated platelets, can bind to the cell surface and participate in platelet aggregate formation. In this study, we show that thrombin activation leads to the rapid and specific association of a large amount of secreted alpha-granular TSP-1 with the actin cytoskeleton. This cytoskeletal association of TSP-1 was correlated with platelet secretion, but not aggregation, and was inhibited by cytochalasin D, an inhibitor of actin polymerization. Association of TSP-1 with the actin cytoskeleton was mediated by membrane receptors, as shown by using MAII, a TSP-1-specific monoclonal antibody that inhibited both TSP-1 surface binding to activated platelets and cytoskeletal association. TSP-1 and its potential membrane receptors, e.g. alphaIIbbeta3 integrin, CD36 and CD47, concomitantly associated with the actin cytoskeleton. However, studies on platelets from a patient with type I Glanzmann's thrombasthenia lacking alphaIIbbeta3 and another with barely detectable CD36 showed normal TSP-1 surface expression and association with the actin cytoskeleton. Likewise, no involvement of CD47 in TSP-1 association with the actin cytoskeleton could be inferred from experiments with control platelets using the function-blocking anti-CD47 antibody B6H12. Finally, assembly of signalling complexes, as observed through translocation of tyrosine-phosphorylated proteins and kinases to the actin cytoskeleton, was found to occur in concert with cytoskeletal association of TSP-1, in control platelets as well as in thrombasthenic and CD36-deficient platelets. Our results imply a role for the actin cytoskeleton in the membrane-surface expression process of TSP-1 molecules and suggest a possible coupling of TSP-1 receptors to signalling events occurring independently of alphaIIbbeta3 or CD36. These results provide new insights into the link between surface-bound TSP-1 and the contractile actin microfilament system which may promote platelet aggregate cohesion.