You are here

Thesis abstracts

Mattia Andreoletti


Clinical trials and drug regulation: a Philosophical inquiry
My research has been organized around two overlapping questions: (1) whether and how the current approach to the regulation of new medicines should be modified to manage issues generated by new biomedical products, and (2) whether this issue may be reduced to an epistemological and methodological analyses. First of all, there is no way to tackle the issue of changing drugs regulation exclusively from an epistemological perspective. Epistemological and methodological considerations are important, but nonetheless the financial and political issues at stake in drug regulation are enormous. Thus, any epistemological analysis taken out of context cannot make any addition to current debates. Secondly, we must be aware that scientific landscape is rapidly and constantly changing. The molecular revolution in biomedicine makes urgent and crucial to reconsider a regulatory system that has been put in place to regulate products which were essentially different. We should as well take very seriously the fears of the potential negative consequences of adopting looser regulatory standards of evidence, especially because so far RCTs have offered a reasonable level of safety to patients. The good news, in our view, is that there is room for changes without losing much safety for population. The bad news is that changes for the better would require a more systematic and collective effort, which might be beyond any political and academic willingness.

Elisa Barbieri


Modulation of stem cell signalling pathways by nucleophosmin and its leukemogenic mutant
About one third of acute myeloid leukemias (AML) is characterized by the aberrant cytoplasmic localization of nucleophosmin (NPM1), a ubiquitously expressed phosphoprotein that acts as a molecular chaperone and shuttles between nucleus and cytoplasm. Although several animal models have been generated to unravel the mechanism of action of the cytoplasmic mutant NPMc+, it remains poorly understood. We identified a novel function of both wild type NPM1 and NPMc+ in the modulation of Wnt signaling during zebrafish development and primitive hematopoiesis. The injection of NPMc+ and human NPM1 mRNAs in one cell stage zebrafish embryos reveals an opposite effect of the two proteins in the modulation of the Wnt signaling: NPM1 can inhibit the pathways whereas the mutant can activate it. Furthermore, NPM1 and NPMc+ have an opposite effect on the expression of dkk1b, a well known inhibitor of the Wnt pathway, and the co-injection of NPM1 and NPMc+ mRNA rescues the phenotype, suggesting a dominant negative effect of the mutant on the wild-type. Through whole mount in situ hybridization, markers of hematopoiesis have been studied revealing that the myeloproliferative effect of NPMc+ can be overcome by the co-injection of dkk1b, suggesting that the mutant can act by activating the pathway. Taken together, data presented in this thesis suggest that NPMc+ can activate Wnt signaling and that the pathway may be involved in the mechanism of NPMc+ AML establishment and/or progression.

Azzurra Cottarelli


Fibroblast growth factor binding protein 1 (FGFBP1) contributes in the nestablishment and maintenance of the Blood Brain Barrier
hat regulates the permeability between the blood and the Central Nervous System, allowing the entrance of nutrients and the exit of toxic metabolites. This specialization of the brain microvasculature results from the interaction of the endothelium with other components of the NeuroVascular Unit (NVU): Basement Membrane (BM), pericytes and astrocyte end-feet. The canonical Wnt/β-catenin pathway regulates BBB initiation and maintenance. Affymetrix analysis demonstrated that, upon Wnt3a stimulation in murine primary ECs isolated from brain (bMEC), one of the most upregulated transcripts is that of Fibroblast Growth Factor Binding Protein 1 (FGFBP1), a cargo protein that, after secretion in the extracellular matrix (ECM), binds the FGF immobilized in the ECM and mobilizes it, protecting it from degradation and presenting it to FGF tyrosine-kinase receptor. Since Wnt3a stimulation selectively induces FGFBP1 expression in brain ECs, we hypothesized that it could be involved in the process of initiation and/or maturation of the BBB. We demonstrated in vivo in zebrafish that FGFBP1 knock down by morpholino presents vascular abnormalities in the brain, together with cerebral hemorrhages and impaired permeability. Using an endothelial-specific FGFBP1 knock out murine model, we further demonstrated that inhibition of endothelial FGFBP1 expression affects brain vascular development, causing vascular defects and increased BBB permeability and influencing the number of pericytes and the composition of the BM. Finally, in vitro FGFBP1 absence promoted a “tip-like” phenotype and an increase in the expression of Plvap in bMECs. In conclusion, our work proposes a novel role for FGFBP1 in the maintenance of the properties of the BBB and in the regulation of the complex intercellular interactions occurring within the NVU.

Giuseppina D'Alessandro


The role of RNA and DNA:RNA hybrids at DNA double-strand breaks
The stability of our genome is constantly challenged by several genotoxic threats. DNA double-strand breaks (DSBs) are the most dangerous DNA lesions that, if not repaired, can lead to cancer initiation and progression and/or ageing. These detrimental consequences can only be avoided if cells promptly recognize the lesions and signal their presence, thus promoting either efficient repair and transient cell cycle arrest or cell death and cellular senescence. This is the role of the DNA damage response (DDR) proteins and the newly identified damage-induced non coding RNAs. We recently discovered that RNA polymerase II is recruited to DSBs and synthetizes damage-induced non-coding RNAs (dilncRNAs). DROSHA- and DICER-mediated processing of dilncRNAs generates small RNA species, named DNA damage response RNA (DDRNAs) (Francia, 2012), that localize to DSBs via pairing with dilncRNAs and promote DDR signaling (Michelini et al., in press). Similar small non-coding RNA species discovered in plants are involved in DNA repair by homologous recombination (HR) (Wei, 2012, Gao, 2014, Wang, 2016). In line with these results, I report that transcriptional inhibition impairs recruitment of the HR proteins BRCA1, BRCA2, and RAD51 to DSBs, while partially promoting DNA end resection. Moreover, I show DNA:RNA hybrids accumulation at DSBs in mammalian cells by both DRIP analyses and imaging techniques. Damage-induced DNA:RNA hybrids form upon the hybridization of RNA species, likely dilncRNAs, to the resected DSBs DNA ends generated during the S/G2 cell cycle phase. I also report that purified recombinant BRCA1 binds DNA:RNA hybrids in vitro; moreover, DNA:RNA hybrids in vivo contribute to BRCA1 recruitment to DSBs. Consistent with the need to tightly regulate DNA:RNA hybrid levels, I demonstrate that RNase H2, the major RNase H activity in mammalian nuclei, is recruited to DSBs through direct interaction with RAD51. In summary, I report for the first time that DNA:RNA hybrids accumulate at DSBs in mammalian cells in a cell-cycle- and DNA end resection-depended way. At DSBs, BRCA1 directly recognizes DNA:RNA hybrids and likely controls their turn-over by mediating the recruitment of RNase H2 via RAD51.

Luana D'Artista


The role of Pin1 in Myc induced lymphomagenesis
The c-myc proto-oncogene is activated by translocation in Burkitt's lymphoma and is frequently subject to secondary mutations. Substitutions in codon 58, in particular, stabilize the c-Myc protein (Myc) and augment its oncogenic potential. In wild-type Myc, phosphorylation of Ser 62 primes phosphorylation of Thr 58, providing a landing pad for the peptidyl prolyl-isomerase Pin1, which in turn promotes Ser 62 dephosphorylation and Myc degradation. However, genetic data are missing to address whether Pin1 influences Myc-induced lymphomagenesis. We observed that genetic ablation of Pin1 in Eμ-myc transgenic mice starkly reduced lymphoma onset and penetrance. In pre-malignant Pin1 deficient B-cells, the proliferative response to Myc was selectively impaired, in the absence of changes in either steady-state Myc levels or Ser 62 phosphorylation. Similarly, in Pin1-/- mouse embryo fibroblasts (MEFs), prolonged Myc activation inhibited proliferation. In either B-cells or MEFs, Myc-induced apoptosis occurred independently from Pin1. Loss of p53 rescued proliferative defect of Eμ-myc Pin1-/- B cells and of Pin1-/- MEFs upon Myc activation. This result suggests the existence of a p53 dependent checkpoint in the Pin1-/- cells that inhibits Myc driven proliferation: this mechanism blocked tumorigenesis in the E-myc Pin1-/- mice. The origin of this arrest is under investigation. Although Eμ-myc Pin1-/- B cells showed reduced activation of Mycinduced mRNAs relative to Eμ-myc Pin1+/+ cells, experiments in MEFs and R26MycER B cells demonstrated that the overall transcriptional activity of Myc was independent from Pin1. Thus, while Pin1 is not essential for normal cell growth and mouse development, it is required to support the oncogenic activity of Myc in B cells. Moreover, Eμ-myc lymphoma cells were sensitive to Pin1 inhibition, making Pin1 an attractive therapeutic target in Myc-driven tumors.

Claudia Iavarone


Functional characterization of the endocytic protein Epsin3 in breast cancer
Endocytosis plays a critical role in the maintenance of cellular homeostasis. One relevant, unanswered, question regarding the role of endocytosis in the cell is whether subversion of the endocytic route is involved in the development of pathological conditions, such as cancer. This possibility is supported by studies showing altered expression of several endocytic proteinsin human tumors. In a previous study in our lab, the endocytic protein Epsin3 (Epn3) was found in a gene signature prognostic for metastatic breast cancer. Epn3 belongs to the Epsin family of endocytic proteins. Unlike the other Epsin members, which are ubiquitously expressed, Epn3 is exclusively expressed in gastric cells and in wounded or pathological tissues. While Epn1 and Epn2 have been well characterized as endocytic adaptors, the exact function of Epn3 protein in endocytosis is largely unknown. In the present study, we show that Epn3 is overexpressed in breast tumors and that its up regulation correlates with clinical parameters of aggressive disease. We demonstrate that breast tumor cells harboring Epn3 amplification are dependent on Epn3 deregulation for the maintenance of their tumorigenic potential. Furthermore, overexpression of Epn3 in breast tumor cells increases the tumorigenic potential in vivo. Of note, Epn3 overexpression is able to induce transcriptional and morphological changes typical of an epithelial-to-mesenchymal transition (EMT), in a TGFβdependent manner. Finally, we demonstrate that Epn3 is involved in Ecadherin Internalization, by inducing its down regulation from the cell surface upon TGFβ-stimulation. In conclusion, our data suggest a novel oncogenic role for Epn3 in Human breast cancer.

Elisabetta Incorvaia


Insight from AID-induced DNA damage resolution: cellular context matters
High-fidelity DNA repair are nuclear mechanisms essential to maintain genome integrity. However those mechanisms can be channeled away from DNA stability towards mutations and recombinations. During immunoglobulin locus diversification in B cells, Activation Induced Deaminase (AID) triggers, through DNA deamination of cytosine into uracil, DNA instability. AID-induced lesions can be resolved via error-free or error-prone DNA repair, depending on cellular milieu and local chromatin context. To uncover the regulation of repair pathway choice, AID damages on a DNA plasmid were resolved using cellular extract either via base excision repair (BER: short patch SP-BER or long patch LP-BER) or mismatch repair (MMR). Cell-type origins presented quantitative differences in DNA repair kinetics: a) overall sensitivity, b) non-B cells activating non-canonical MMR first, c) B cells activating SP-BER first, and d) LP-BER only activated in B cells. Analysis of single components known to influence DNA repair, such as chromatin and nucleosome formation, revealed significant changes in DNA repair pathway choice. DNA with nucleosome favourable base-stacking preferred LP-BER, while non-nucleosome stacking DNA preferred SP-BER and MMR. Overall these results provide insight into the cellular context that can influence DNA repair. The use of B cells and cancer cell lines recapitulates in vivo Ig locus diversification, while at the same time provide insight into the mechanisms of tumourgenesis.

Silvia Masella


3D genomic organization of mouse macrophages
The spatial organization of the genome and its biological function are intimately linked. It is becoming evident that transcription regulation often involves multiple long-range regulatory elements and it’s influenced both by the genomic environment and by the shape of the genome. Recent studies based on Chromosome Conformation Capture-derived techniques, showed that eukaryotic cells organize their chromosomes into topological domains that are largely invariant among cell types and where the majority of looping interactions between regulatory elements take place (Dixon et al., 2012; Nora et al., 2012; Sexton et al., 2012). The principles defining the relationships between these elements and distal target genes remain poorly understood. Previous studies lack either the spatial resolution or the temporal coverage to observe possible dynamic changes in chromatin contacts between promoters and their corresponding distal regulatory elements during gene activation. Here we exploited high-resolution 4C and 5C techniques to elucidate principles of 3D organization of the macrophage genome, in both basal conditions and after macrophage activation, dissecting the specific role of the macrophage master regulator PU.1 in the formation of the general chromosome topology. Our findings indicate that the global organization of chromatin contacts is to a large extent unaffected by macrophage activation, which only partially impacts the looping between specific regulatory elements. Our analysis also reveals an high cell-type specificity of macrophage promoter-enhancer interactions, which is not dependent on the presence of PU.1.

Maria Pelliccia


Strategies for enhancing viral gene transfer and the thermostability of viral vectors in vaccine applications
Viruses represent powerful gene delivery systems in biomedicine. For successful gene therapy and immunization programs, the efficiency and stability of viral vectors are fundamental aspects (Jorio et al. 2006). To address this challenge, in the present project we investigated the interaction between viruses and nanomaterials. We tested three types of sulfonate- functionalized gold nanoparticles (AuNPs), MUS:OT, MUS and MUS:brOT NPs, which are small than 5 nm, negatively charged and poorly cytotoxic (Verma et al. 2008). As virus models we focused on human recombinant adenovirus type 5 (Ad) and on two enteroviruses, echovirus-1 (EV1) and coxsackievirus B3 (CVB3). We carried out two sets of experiments: (1) Short-term pretreatment of Ad with nanomaterials for 1 h at 37°C that showed a significant increase in the gene expression in vitro and in vivo. The NPs- enhanced adenovirus transduction aims to reduce Ad vector doses in vivo in order to minimize the adverse reactions of the immune response due to high vector dosage. (2) Long-term thermostabilization studies of Ad, EV1 and CVB3 in vitro in the presence of nanomaterials and other compounds such as sugars at 37°C or room temperature for extensive periods of time. Nanomaterials and sucrose increased substantially the heat stability of viruses. To elucidate the thermal inactivation mechanism of viruses and the stabilizing effect observed, we developed an analytical theory. This research fits in the context of developing more thermo-stable preparations for vaccine that do not require the challenging cold chain in order to preserve the effectiveness of viral vaccines.

Elena Zagato


Role of bacteria and the mucus system in intestinal tumorigenesis
Studies on human colorectal cancer (hCRC) samples have documented a dysbiosis associated with the tumor, but a clear picture of the microorganisms whose abundance is altered during has not yet emerged. This study is aimed at dissecting the role of bacteria in CRC development by focusing on the intestinal mucus barrier as a key mediator in microbiota-host interaction. We used the ApcMin/+ murine model to demonstrate that in tumor-bearing mice the mucus and the intestinal barrier have altered properties. Mucin expression is altered at the level of dysplastic crypts, strictly relating mucus changes with neoplastic transformation. We further observed that bacterial penetrance was compromised in tumor-prone mice with increased Salmonella penetration in the intestine and mesenteric lymph nodes. To address the potential role of mucus alteration in the tumorigenic process we exogenously modified the mucus barrier. Mucus disruption in both the ApcMin/+ and the AOM/DSS models led to increased colonic tumorigenesis. We did also investigate whether there was a dysbiosis associated with tumor progression. Metagenomic analysis in the faeces of mice at different ages highlighted a dysbiosis already at 4 weeks of age in ApcMin/+ mice when tumors are not yet established. In particular, Lactobacillus genus was expanded in ApcMin/+ mice concomitantly with a contraction in the Clostridium genus. Finally since epigenetic mechanisms have been hypothesised to contribute to the loss of heterozygosity of the normal apc allele and bacteria can alter miRNA expression we analysed miRNA profiles in WT and ApcMin/+ intestinal tissue observing alterations in ApcMin/+ ilei.

Marek Adamowicz


NOTCH1 inhibits the DNA damage response by impairing the formation of the ATM-FOXO3a-KAT5 complex
Eukaryotic cells have evolved the ATR/hCHK1, MEC1/RAD53 kinase-mediated signal transduction pathway, known as replication checkpoint, to protect and stabilize stalled replication forks in human cells and budding yeasts, respectively. rad53 mutants, exposed to high doses of the DNA replication inhibitor hydroxyurea (HU), accumulate hemireplicated, gapped and reversed forks, while treatments with low HU doses induce massive chromosome fragmentation. The aim of my work was to better understand the molecular mechanisms through which Rad53 prevents unusual alterations of the architecture of the stalled replication forks and chromosome fragility, under replication stress. We revealed that Rrm3 and Pif1, DNA helicases assisting fork progression across pausing sites in unperturbed conditions, are detrimental in rad53 mutants experiencing HU-induced replication stress. Rrm3 and Pif1 ablation synergistically rescues cell lethality, chromosome fragmentation, replisome dissociation, fork reversal and ssDNA gaps formation at the forks of rad53 cells exposed to replication stress. We provide evidence that Pif1 and Rrm3 associate with stalled DNA replication forks and are regulated through Rad53-mediated phosphorylation.  Our findings uncover a new replication-stress-induced regulative loop in which Rad53 down regulates the Pif1 DNA helicases at the stalled replication forks. In the second part of this thesis we examined the crosstalk between Rrm3, Pif1, the mediator of the DNA damage checkpoint Rad9 and the nuclease Dna2, during unperturbed DNA replication. The experimental evidence collected in this second part of the project, together with pioneering work previously reported from other laboratories, strongly suggests that Dna2, Pif1 and Rrm3 cooperate to finalize late stages of DNA replication.

Julio Aguado


The Role Of Telomeric RNA At Dysfunctional Telomeres And Its Impact On Senescence And Aging
A novel class of small non-coding RNAs discovered in our laboratory, termed DNA damage responseRNAs (DDRNAs), has been demonstrated to be generated upon DNA double strand break (DSB) induction, and to be necessary for full DNA Damage Response (DDR) activation. DDRNAs are generated following DSB induction upon transcription of the damaged locus and the synthesis of an RNA precursor further processed by the endoribonucleases DICER and DROSHA. The aim of this PhD dissertation was to investigate the mechanism underlying DDRNA-dependent DDR activation specifically at telomeres, important chromosomal regions required for genomic stability that, if disrupted, are associated with aging-related diseases. In this dissertation, I show that telomere dysfunction, like DSBs, induces the transcription of telomeric DDRNAs and their precursors from both DNA strands of the telomere. Such transcripts are necessary for DDR activation and maintenance at dysfunctional telomeres. Most importantly, the use of sequence-specific antisense oligonucleotides (ASOs) allows the inhibition of telomere transcripts’ functions, thereby specifically inhibiting telomeric DDR. Telomere dysfunction is rising as a key feature in Hutchinson–Gilford Progeria Syndrome (HGPS) and other premature aging syndromes. Here I show that progerin, the protein whose expression causes HGPS, induces the transcription of telomere transcripts, both in vitro and in vivo. Furthermore, signaling inhibition of progerin-driven telomere dysfunction improves the growth potential of progerin-expressing cells. Finally, this inhibition also increases the lifespan of an HGPS mouse model, opening the possibility for the use of this approach as a viable therapy to treat HGPS.

Arta Ajazi


Atg6/Beclin 1 couples the replication stress response to amino acid metabolism
In the budding yeast Saccharomices Cerevisiae, Atg6 is a non-­‐catalytic component of the phosphatidylinositol (PtdIns) kinase complex Vps34-­‐Vps15-­‐Atg6, which phosphorylates  PtdIns  to  produce  phosphatidylinositol  3-­‐phosphate (PtdIns(3)P). Atg6 is conserved among species, including its mammalian ortologue, Beclin 1, which is partially inactivated in breast and ovarian cancers. Beclin 1 involvement in human cancer is not fully established, but it could act as a tumor suppressor by regulating autophagy.     Genome instability is a hallmark of cancer cells. The replication stress checkpoint has evolved to prevent the occurrence of genome instability. Once activated, the checkpoint orchestrates series of protective responses, including the production of deoxyribonucleotides (dNTPs) and DNA repair.   In this work we show that low levels of intracellular PtdIns(3)P, as caused by deletion of ATG6,  protect cells during replication stress conditions. Moreover, sensitivity to replication stress is enhanced in the presence of amino acid imbalances in the extracellular medium. This effect depends on PtdIns(3)P involvement in endosomal vesicle trafficking, but is independent from both autophagy and the canonical replication stress checkpoint mediated by Rad53. PtdIns(3)P levels, as determined by the activity of the Vps34-­‐Vps15-­‐Atg6 on endosomal membranes, likely affect the internalization of amino acids that are crucial  to  survive  during  replication  stress.  Elucidating  the  molecular mechanisms that link the metabolism of specific amino acids with the response to  replication  stress  will  have  profound  impact  in  understanding  the  role  of Beclin 1 in human cancer and also in cancer therapy.

Federica Alberghini


The role of PRC1 in B cell development and adaptive immune responses
Polycomb (PcG) proteins are epigenetic modifiers that modulate accessibility of genomic loci through the sequential activity of at least two functionally and biochemically distinct complexes, named Polycomb repressive complex 1 and 2 (PRC1 and PRC2), which modify histone tails to yield permanent and heritable, yet reversible, locus silencing. PcG proteins modulate expression of genes involved in multiple biological functions and their deregulation leads to aberrant differentiation of a number of cellular lineages, including the lymphoid subsets. Using a conditional gene targeting approach in vivo, we addressed the function of Ring1a and Ring1b, the catalytic subunits of PRC1, in B cell development. Selective PRC1 inactivation in transitional B cells led to aberrant differentiation and extensive transcriptional deregulation. Loss of PRC1 in germinal center (GC) B cells caused a significant reduction in their numbers and frequency. Consequently, serum titers of antigen-specific, class-switched antibodies were significantly decreased and memory B cell formation was impaired in mutant mice. Instead, mutant GC B cells showed premature onset of terminal differentiation towards the plasma cell fate and increased apoptosis. Importantly, activation of B cells with a non-Activation Induced Deaminase (AID)-inducing compound normalized the apoptotic rate of PRC1 deficient cells to wild-type levels. Together with the up-regulation of several targets belonging to the network of the GC master regulator Bcl6, this suggests that PRC1 may support GC function by co-operating in the establishment of Bcl6 transcriptional program and by participating in the repair of AID-induced DNA damage, allowing tolerance of AID genotoxic activity by GC B cells.

Wondimu Alemu Gebreyes


An electrochemical microfluidic biosensor platform fabricated by additive manufacturing and supersonic cluster beam deposition
Fused Filament Fabrication (FFF) three-dimensional printing have attracted much attention for fabrication of microfluidic platforms used to construct electrochemical microfluidic biosensors because of high process speed, low production costs and the possibility of manufacturing directly from virtual data. Because of poor adhesion between metal electrodes fabricated using conventional techniques and FFF printed thermoplastic substrates, electrodes are usually integrated into the devices either modularly or using adhesive layers placed at the bottom of fluidic channels. These have hindered the exploitation of FFF for scale-up manufacturing of monolithically integrated microfluidic biosensors. In this work, supersonic cluster beam deposition (SCBD) was employed to fabricate strongly anchored nanostructured electrodes integrated into FFF printed microfluidics platforms. SCBD enables the formation of well-adhering metallic thin film electrodes by implanting supersonically accelerated neutral metal clusters into polymeric substrates. The SCBD also enables deposition over large areas using noble metals and metal oxides with precisely controlled geometry and surface topography. A novel integrated manufacturing approach was developed and optimized to couple SCBD fabricated electrodes with consumer-grade FFF printed microfluidics, employing acrylonitrile butadiene styrene as the base material, to develop a three electrodes configuration electrochemical sensor on-a-chip. Electrochemical investigation performed using stagnant ferro/ferricyanide probe showed that the integrated device possesses high sensitivity and functionality as an electrochemical sensor. In addition, in-channel laminar flow electrochemical detection conducted using the same probe showed robust stability in the system response for online dynamic detection. The integrated platform could be employed for various customized clinical, industrial, and environmental sensing applications.

Gabriele Alfarano


Interferon Regulatory Factor 1 (IRF1) links immunological and metabolic traits to histological grade in pancreatic cancer
Pancreatic Ductal Adenocarcinoma (PDAC) is the most frequent neoplasia of the exocrine pancreas with a 5-years overall survival of less than 5% . Poor response to treatments can be attributed, at least in part, to the pervasive heterogeneity of this type of cancer, whose histological differentiation grade ranges from a well differentiated to a poorly differentiated phenotype . My lab previously dissected the transcriptional and epigenetic networks underlying PDAC grading. By using cell line models mimicking different PDAC grades, transcriptional data highlighted an interferon-related signature as peculiar of low-grade PDACs. This project aims at investigating the links between this interferon-related signature and the epithelial phenotype of well-differentiated PDACs. The role of Interferon Regulatory Factor 1 (IRF1), a transcription factor critical for the interferon response, in PDAC differentiation will be investigated. By combining a loss-of-function strategy and RNA-seq this work aimed at defining the targets and the effects of IRF1 differential expression in the mentioned cell line model. Model reliability will also be confirmed via immunohistochemistry on tumour microarrays. The immunological and metabolic phenotypes regulated by IRF1 in well differentiated PDACs will further be analysed by the use of xenografts and cell culture based assays. It will be shown that IRF1 regulates multiple interferon related genes involved in antigen processing and presentation; its deletion affects stromal composition in xenografts. IRF1 deficient cells also rewire their metabolic networks, with changes in lipid composition and metabolite usage. Overall, we found that the transcription factor IRF1 pleiotropically controls a variety of phenotypical traits of low grade PDACs: from antigen processing and presentation pathways, to metabolism, which IRF1 deficiency rewired towards an increased respiratory and lipogenic profile.

Victor Alfred


Genetic screening to identify interactors of ESCRT-II subunit, VPS25, and preliminary characterisation of candidates
ESCRT (Endosomal Sorting Complex Required for Transport) proteins regulate cell surface receptor degradation by sorting and packaging ubiquitinated cargoes into the intraluminal vesicles of multivesicular bodies (MVBs). A range of human diseases including cancer, and neurodegeneration display altered expression or are caused by mutations of ESCRT subunits. Studies have shown that Drosophila tissues lacking ESCRTs display neoplastic-like features like overproliferation and polarity defects, partly due to aberrant signalling including Notch signalling. To understand ESCRT-regulated processes in vivo, we utilised modification of a deformed wing phenotype specifically caused by knockdown (RNAi) of Vps25, an ESCRT-II subunit. We systematically screened chromosomal regions and identified 204 genetic interactors of Vps25 that enhanced/suppressed the phenotype. They include genes that function in trafficking, signalling, transcription, ion transport and many other biological processes; suggesting that ESCRTs influence a wide range of biological processes. We have focused on a subset of these hits that regulate tissue growth with a secondary screen based on modification of a Delta-driven eye overgrowth phenotype, isolating a subset of 43 genes involved in regulating tissue growth, some of which are novel and uncharacterised. Human orthologues of some of these genes are important in cancers; dropout (dop), whose mammalian orthologues are the MAST kinases, have been shown to contribute towards breast cancer development. dop mediates Delta-driven eye overgrowth possibly by upregulating Delta expression. In human cells, MAST2 does not affect Notch signalling but might contribute to tumorigenesis by regulating the NFκB pathway. We have also characterised another interactor, CG12163 which is the homologue of mammalian Cathepsin F. Mutations in Cathepsin F cause a rare form of neuronal ceroid lipofuscinosis (NCL) called Type B Kufs disease. Our Drosophila model which recapitulates aspects of the human disease phenotype suggests that defects in autophagy might underlie the pathogenesis of NCLs.

Letizia Amadori


Identification of molecular mechanisms responsible for degradation of the tumor suppressor protein NUMB in cancer
NUMB was initially described as a cell fate determinant involved in neurogenesis. More recently, NUMB has been implicated in different types of human cancers, in which it has a tumor suppressor role. My research has been focused on the identification of the molecular players responsible for NUMB degradation, which is due to exaggerated ubiquitination and could be restored by proteasome inhibition by MG-132, in breast and lung tumors. We have assessed an siRNA high-throughput ELISA-based screening for detection of NUMB protein increase upon gene silencing in a NUMB-deficient breast cancer cell line, MDA-MB-361. We have started screening a library containing all human E3-ligase families first and, after two rounds of screening, we validated RBX1 and FBXW8 proteins as putative NUMB downregulators both in breast and lung tumors from human patients, in which NUMB basal expression is low respect to their normal counterpart and could be restored to normal levels after both RBX1 and FBXW8 silencing. These results have opened the possibility that the mechanism responsible for NUMB degradation in NUMB-deficient tumors is a cullin-RING ligase complex-dependent one. We have also demonstrated that the in vivo treatment of immunocompromised mice with the proteasome inhibitor MG-132 inhibits the growth of NUMB-deficient breast tumors and not that of NUMB- proficient ones through the restoration of NUMB protein levels. This evidence pointed out that targeting loss of NUMB could be used as an anti-cancer therapy for NUMB-deficient breast tumors. We are also performing experiments to demonstrate that the silencing of RBX1 and FBXW8 could inhibit NUMB-deficient tumors' growth potential as well.

Omer An


Role of somatic copy number variations in cancer
Genetic variation is the main reason of the phenotypic differences among individuals, as well as of many human genetic diseases. Recent advances in the methods to study the human genetic variation allow better identification of its different forms, in particular of copy number variations (CNVs). The causative role of germline CNVs in Mendelian diseases and in cancer predisposition is well established. Moreover, the driver role of cancer somatic CNVs is recently emerging, and large-scale quantitative analyses elucidating their functional role in cancer genomes are needed. To achieve this, we have analysed the genomic landscape of somatic CNVs in cancer genomes in comparison to germline CNVs in the genomes of healthy individuals. We observed that somatic CNVs substantially affect the genic portion of the genome, preferentially targeting cancer genes. Moreover, this is independent of genomic features, such as DNA repeating elements and recombination rate. In particular, we confirmed that oncogenes are preferentially amplified and tumour suppressors are preferentially deleted. To investigate their functional impact, we measured the gene expression changes upon copy number variation. We observed that amplification of a gene leads to its higher expression whereas deletion results in decreased gene expression, which suggests that amplifications activate dominant genes and deletions inactivate recessive genes. The two classes of cancer genes are vastly modified consistent with their functional roles as oncogenes and tumour suppressors, with the few exceptions of frequently amplified recessive genes underlying complex epigenetic regulation. The mutational spectrum of the human genes in cancer, together with their systems level properties, can be exploited to identify novel targets for anti-cancer therapy, in which synthetic lethality emerges as a promising approach. Based on the working hypothesis that paralogous genes may engage in synthetic lethal interactions due to the functional redundancy between them, we combined several gene properties to predict synthetic lethality between paralogous gene pairs. Out of 37 candidate gene pairs, we experimentally validated the synthetic lethal interaction between two components of the cohesin complex, STAG1 and STAG2. Finally, we present the latest release of Network of Cancer Genes (NCG 5.0), a manually curated database of cancer genes and their systems-level properties. NCG 5.0 collects a list of 1,571 cancer genes mutated in 13,315 cancer samples and 24 primary sites from 175 published papers. NCG has been increasingly appreciated as a central resource for cancer genomics research, facilitating candidate prioritization for hypothesis testing and experimental planning in a wide range of studies.

Francesca Angiolini


The adhesion molecule L1: a novel player in ovarian cancer vasculature
Ovarian cancer (OC) represents an outstanding clinical challenge because of its high mortality rate, due to tumor relapse and chemoresistance. The identification of novel strategies for the treatment of OC is clearly an unmet clinical need. In this context, drugs interfering with tumor neovascularization have shown promising results in clinical trials, however, with only transient beneficial effects. Thus, the definition of novel druggable targets within OC vasculature will likely contribute to improve the clinical management of patients. L1 is a transmembrane adhesion molecule in the nervous system. However, several studies have demonstrated its involvement in different types of human cancer. In this context, L1 expression is generally associated with poor diagnosis, aggressive behavior and increased metastatic spread. Our laboratory has obtained compelling evidence that L1 is aberrantly expressed in tumor vasculature and exerts a pleiotropic function in endothelial cells. Based on these findings and on the pivotal role of angiogenesis in ovarian cancer, in this work I have investigate the functional role of L1 within the OC-associated vasculature. My results revealed that L1 is abundant in OC vasculature as compared to normal vessels and it was found to be a causal player in OC progression. This research, besides giving insights into novel pathways involved in pathological angiogenesis, provides the rationale for exploring the clinical relevance of L1 expression and function in OC vessels and in their crosstalk with tumor cells, possibly opening new avenues for the development of innovative targeted therapies for OC malignancy.

Marco Annoni


Lie to Me
. The Ethics of Truth-telling and Deception for Oncology
This dissertation deals with the ethics of truth telling and deception in medicine. Should clinicians tell the truth, even if the truth may cause patients irremediable psychological and physical harm? Are lies told with a benevolent intent always blameworthy? Are deception and concealment less culpable than lying? Should patients be informed that the medicine they are taking “with so many benefits” is just a placebo? How can physicians determine, between two alternative verbal disclosures, which one provides the optimal balance between truthfulness and beneficence? In this dissertation I endeavor to elaborate an answer to these and other questions by setting forth a normative theory of truth telling and benevolent deception for medical professionals, and especially for those operating in clinical oncology. Throughout this work I defend two main ideas. First, clinicians have a duty of veracity in all their professional communications, but in exceptional cases other considerations of beneficence and compassion may override this prima facie obligation. Second, clinicians may resort to clinical deception only if they have ruled out all other truthful courses of action and would be ready to hypothetically defend and actually disclose their behavior in public. This view, I contend, has several advantages over competing accounts and provides clinicians with a practical way of approaching moral dilemmas about truth telling and deception in clinical medicine.

Sina Atashpaz


A cell reprogramming-based approach to study 7q11.23 gene dosage imbalances in Williams Beuren syndrome and autism spectrum disorder
Symmetrical gene dosage imbalances at 7q11.23 chromosomal region cause two unique neurodevelopmental diseases, Williams Beuren Syndrome (WBS) and the 7q11.23 microduplication associated to autistic spectrum disorder (7dup-ASD). Although both these diseases share common features such as intellectual disability and craniofacial dysmorphism, they can be distinguished by distinct social and language abilities: WBS patients characterized by hypersociality and comparatively well-preserved language skills while 7dup-ASD is associated with impairment in social interaction and communicative skills. The involvement of same genetic interval in these disease, points out to small subset of dosage-sensitive genes affecting cognition, social behavior and communication skills. Among the genes in the deleted region, some were shown to contribute to the abnormalities in these patients through transgenic mice models and individual case reports. However, the precise cellular and molecular phenotypes associated with these syndromes in disease-relevant cell-types are unknown due to the scarce availability of primary diseased tissues. Transcription factor induced somatic cell reprogramming has bypassed such fundamental limitation and has enabled us to model human diseases, elucidate their pathogenesis and discover new therapeutics by screening small chemicals/drugs on these models. During my PhD studies, I focused on the functional dissection of these complementary diseases at the level of transcriptional deregulation in patient-derived iPSC and its differentiated derivatives such as neural crest stem cells, mesenchymal stem cells, and neural progenitors. To this end, we have assembled a unique cohort of typical WBS, atypical WBS (patient with a partial deletion) and 7dup-ASD patients (along with unaffected relatives), and then I used mRNA reprogramming to establish and characterize at least 3 independent iPSC lines from a total of 12 individuals. High throughput mRNA sequencing on iPSC revealed critical transcriptional derangements in disease-relevant pathways already at the pluripotent state. These alterations found to be selectively amplified upon differentiation into disease-relevant lineages, thereby establishing the value of large iPSC cohorts in the elucidation of disease-relevant developmental pathways. Finally, we created an open-access web-based platform to make accessible our multi-layered datasets and integrate contributions by the entire community working on the molecular dissection of the 7q11.23 syndromes

Giorgio Baldi


Identification and characterization of stressed replication fork intermediates
The DNA replication machinery can encounter many obstacles while duplicating the genome, which put the replicative apparatus under stress. Cells have developed numerous mechanisms to overcome replication stress (RS). After the loss of key factors, that help to deal with RS, deleterious events can occur. The main pathway implicated in DNA damage response is DNA damage checkpoint. When DNA replication is blocked, checkpoint activation ensures structural stability of the replisome avoiding fork collapse and promoting DNA replication resumption. One of the hallmarks of RS is single-stranded DNA accumulation at the replication forks. Here we used AFM and EM to provide a detailed characterization of DNA lesions arising in RS conditions imposed by DNA polymerase inhibitors in Xenopus laevis egg extract. We identified different intermediates induced by RS, including ssDNA gaps and reversed forks (RVFs). Importantly, we directly correlated the presence ssDNA gaps with the onset RVFs. We identified one possible source of ssDNA gap accumulation at forks by showing that homologous recombination protein Rad51 is required for optimal function of Pola at stressed replication forks. Rad51-Pola interaction is likely to be important for stalled fork resumption. We also provided evidence that replication fork with ssDNA gaps are converted into RVFs by Smarcal1. We also showed RVFs can trigger Mre11-dependent DNA degradation upon RS in the absence of Rad51. We provided mechanistic insights into checkpoint regulation of RVFs levels through ATR, Smarcal1 and Rad51 regulation. Overall this provides structural and molecular insights into the metabolisms of replication forks under stressful conditions.

Elisa Barbieri


Contact sites between the endoplasmic reticulum and the plasma membrane control EGFR endocytosis
The epidermal growth factor receptor (EGFR) can be internalized through different routes. While clathrin-mediated endocytosis destines EGFR for recycling and signaling, internalization through non-clathrin endocytosis (NCE) targets the receptor for degradation. Since NCE appears to be a major negative regulator of EGFR levels, a more complete picture of this pathway would likely reveal new insights into aberrant EGFR signaling observed in many types of cancer. By combining a candidate gene approach with an unbiased proteomic approach, we have defined EGFR-NCE as molecularly distinct from other NCE pathways, relying on functional regulators not previously implicated in endocytosis. We found that reticulon 3 (RTN3), an endoplasmic reticulum (ER)-resident protein, is fundamental for NCE-mediated EGFR internalization, and that its ablation delays EGFR degradation, demonstrating that the NCE pathway is a critical regulator of the EGF-dependent cellular response. We show that, upon stimulation with high dose of EGF, RTN3 is localized in close proximity to EGFR and that it is crucial for the formation of contact sites between the ER and the plasma membrane (PM), which are needed for NCE to proceed efficiently. We also show that ER contact sites are involved in local Ca2+ release: high EGF doses induce a release of Ca2+ from the ER to the PM, which is strongly inhibited upon knockdown of RTN3. This calcium release depends on the inositol trisphosphate (IP3) cascade and is essential for the internalization of EGFR via NCE.  

Iros Barozzi


Overlapping sequence features of mammalian enhancers coordinately control engagement of transcription factor consensus sites and nucleosomal occupancy
In mammalian cells transcription factors (TFs) bind only to a small fraction of the available consensus sites in the genome. In particular, they prefer sites embedded in regions of computationally predicted high nucleosomal occupancy. This is compatible with non-exclusive mechanisms of nucleosome-driven TF-binding and nucleosome-mediated masking of TF binding sites, suggesting that TFs, and in particular pioneers, must overcome a strong barrier in order to engage binding. Exploiting the available information for the hematopoietic master regulator Pu.1, we applied machine-learning approaches and uncovered the sequence-encoded information that discriminates engaged from non-engaged genomic consensus sites. We identified a minimal set of features, which predicts Pu.1 binding with 78% accuracy, among which sequence determinants able to drive nucleosome occupancy were found. Consistent with this, while Pu.1 maintained nucleosome depletion at many thousand cell type-specific enhancers in macrophages, these site are otherwise occupied by nucleosomes in other cell types and in in vitro reconstituted chromatin. As predicted, engaged consensus sites showed higher sequence-encoded nucleosome occupancy compared to the myriad of non-occupied (and likely non-functional) consensus sites that randomly occur in mammalian genomes. The same sequence features selected in machine learning also explains up to 45% of the variability observed in the nucleosome occupancy in cells where Pu.1 is not expressed (a performance equal or better than what achieved by ad hoc models), suggesting that the same information contributes to nucleosome occupancy and positioning. These data reveal a basic organizational principle of mammalian enhancers whereby TF-engagement at its consensus sites and nucleosome occupancy are coordinately controlled by overlapping sequence features. This model also suggests that co-evolution of these features may be crucial to ensure cell-type specific enhancer activation. The nucleosomal patterns at Pu.1-bound sites in macrophages were further characterized, uncovering distinct subtypes with different DNA sequence composition, which mirror distinctive nucleosomal configurations either in the presence or in the absence of Pu.1.

Federica Basilico


A biochemical and structural study of the kinetochore - centromere interface
Faithful chromosome segregation during mitosis requires the dynamic interaction between spindle microtubules and kinetochores, multiprotein complexes built on centromeres. A group of kinetochore proteins associates with centromeres throughout the cell cycle and is thus named constitutive centromere-associated network (CCAN). Biochemical and functional analyses indicate that CCAN proteins are organized in sub-complexes. However, the exact organization of these sub-complexes has not been fully elucidated to date. The aim of my project has been the biochemical reconstitution of CCAN sub-complexes and their structural and functional characterization. In particular, this dissertation dwells upon the results I have obtained regarding three different but intrinsically related topics. First, I present a biochemical and structural characterization of the CCAN protein CENP-M (centromere protein M), which displays the fold, but not the enzymatic activity of a G protein. In addition, I disclose its unprecedented role in the context of a quaternary complex with CENP-H, CENP-K and CENP-I and provide information about the spatial organization of this complex. The first steps towards an in vivo validation of these results are also described. Second, I report the discovery of a direct interaction of CENP-H / CENP-K complex with CENP-C. Third, I have been involved in establishing in the laboratory techniques for the in vitro reconstitution of recombinant nucleosomes. The production of material of good quality and quantity has recently been achieved, supporting the analysis of in vitro interactions between nucleosomes and kinetochore components. Specifically, I illustrate some preliminary observations concerning a direct interaction of Mis12 complex with nucleosomes.

Giulia Bastianello


The role of ATM in the regulation of cell mechanics
The ATM (Ataxia-telangiectasia mutated) kinase is one of the major players of the DNA damage response and its mutation causes the development of the Ataxiatelangiectasia genetic disorder (A-T). A-T is a rare autosomal recessive neurodegenerative disease characterized by neurological dysfunction, multisystem abnormalities and cancer predisposition. ATM belongs to the family of phosphatidylinositol 3-kinase-related kinases, which also includes ATR, DNA-PK and mTOR. ATM, ATR and DNA-PK collaborate to preserve genome integrity in the face of DNA double strand breaks (ATM, DNA-PK) and replication stress (ATR). Recent observations are expanding the roles of these kinases outside the nucleus and several proteomic screens have identified hundreds of cytoplasmic substrates phosphorylated by ATM and ATR. Accordingly, ATM and ATR are also activated in the absence of DNA damage by multiple stress conditions. Our group previously showed that ATR is activated by mechanical stress and localizes to the nuclear envelope; moreover, recent unpublished observations in our lab suggest that ATR influences cell mechanics regulating nuclear plasticity. Here we investigate the function of ATM in relation to cell mechanics using a variety of techniques to study the mechanical properties of the cells, their plasticity and their ability to migrate across constrictions. We found that lack of ATM activity correlates with structural alterations in the cytosol and increases cell stiffness. Moreover, defective ATM impairs 2D locomotion and interstitial migration. ATM is activated in cells subjected to deformation of the nucleus during the squeezing across constrictions in 3D migration and ATM depletion impairs survival of cells migrating inside constrictions. 12 By mass spectrometry analysis we identified a plethora of proteins involved in the regulation of cytoskeleton and cell structure that physically interact with ATM, many of which are subjected to future investigations. Altogether these observations suggest that ATM plays a role in the control of cell plasticity and migration, possibly through the regulation of cytoskeletal components.

Clara Bernardelli


Different extracellular vesicles subpopulations characterize metastatic progression: qualitative and quantitative analysis of isogenic melanoma cell lines and their secreted factors
The survival and proliferation of metastases is a consequence of the pre-metastatic niche (PMN) evolution, an abnormal, tumor growth-favoring microenvironment devoid of cancer cells. Among tumor derived secreted factors, extracellular vesicles (EVs) are key players in PMN establishment and facilitate organotropic metastasis. Compared to normal melanocytes, melanoma cells produce a large quantity of EVs, that can be detected in the plasma of melanoma patients. For this reason, a full characterization of secreted vesicles subpopulations and of their cargo is necessary to understand how EVs affect PMN formation. In this study, we demonstrated for the first time that EVs secreted by isogenic primary tumor and metastatic melanoma cell lines are quantitatively and qualitatively different, suggesting that diverse EVs subpopulations characterize metastatic progression. We also set a deep quantitative proteomics protocol to analyze the proteome of these cells and of their EVs and soluble secreted factors. WNT5A was found as an important component of primary tumor secreted EVs; on the contrary, we observed a specific APOE and Fibronectin sorting to EVs in in metastasis versus primary tumor cell. Finally, we observed that increased levels of RAB27A protein in metastatic cells do not correlate with an increased EVs secretion. Our preliminary results demonstrate that EVs secreted by RAB27A-KD cells maintain cancer cells clonogenic ability and that low levels of RAB27A expression correlate with higher cells motility. These findings suggest a paracrine activity of a RAB27A -independent EVs subpopulation in tumor-PMN communication to promote cancer progression.

Alice Bertocchi


Role of the Gut Vascular Barrier in metastatic Colon Cancer
Spreading of neoplastic cells from colon cancer to regional lymph nodes (LN) is often associated with distant recurrence. However, a number of clinical trials have shown that lymphadenectomy is not increasing Colon Cancer (CC) patient’s survival. This phenomenon is probably due to a metastatic dissemination that occurs via the systemic blood circulation, rather than the lymphatic vessels. How such dissemination is achieved is unknown. We hypothesized that the disruption of the vasculature in colon tumors can be linked to cancer cells dissemination. We identified a new marker of deranged vasculature which is expressed in primary tumors of CC patients that have developed metachronous distant metastases, independently from the presence of metastases in regional LN. Moreover, we studied vasculature disruption in a mouse model that develops spontaneous tumors mainly in the colon. These mice showed dismantled vasculature at tumor level and they concomitantly exhibited the formation of a pre-metastatic niche at distant sites which favored the recruitment of metastatic cells. Vascular impairment can shed light on the process of colon metastases and this new marker can be used as a prognostic marker for distant recurrence.

Stefania Bertora


Role of nuclear envelope protein MAN1 in nuclear organisation and maintenance of genome stability
The eukaryotic nucleus is characterized by a defined spatial organization of the chromatin, which relies on the physical tethering of many genomic loci to the inner surface of the nuclear envelope, that occurs through lamins and lamin-associated proteins. Man1 is a member of a lamin-associated protein family characterized by the presence of a highly conserved chromatin-binding domain (LEM). Data obtained with yeast Man1 homolog underline the importance of this protein in different processes of the cell cycle, such as chromosome segregation, nuclear pores assembly, gene expression, chromatin organization and maintenance of genomic stability, while in animal models, the function of Man1 has been associated to the regulation of developmental signalling pathways. In this study, truncated mutants of Man1 containing the LEM domain added to Xenopus laevis cell-free extracts were shown to inhibit nuclear assembly and alter nuclear pores formation. Moreover, Xenopus nuclei assembled with Man1 truncated fragments were characterized by defects in chromatin organization, DNA replication and accumulation of DNA damage and, as a consequence, they failed to progress through mitosis.  Furthermore, mouse embryonic stem cells (mESCs) depleted for Man1 showed signs of spontaneous differentiation, indicating inability in the maintenance of stem cell features. Intriguingly, preliminary analysis of Man1-knockout mESCs transcriptional profile showed an alteration of gene expression at the level of pericentromeric and telomeric regions, underlining a potential link between Man1 and genomic stability of these particular regions. In conclusion, this study illustrates the importance of Man1 in ensuring the proper chromatin organization necessary to support different cellular processes.

Ganesh Bhat


Sox17 is a co-effector in Wnt/β-catenin mediated blood-brain barrier development and maintenance
The presence of blood-brain barrier (BBB) at the level of endothelial cells (ECs) that line the capillaries in the brain is a challenge while treating intracranial tumors or many neurological disorders since BBB could block the passage of many solvents or drugs entering from systemic circulation to brain parenchyma. Researchers are trying to understand how BBB is induced during development and maintained in adult life in order to develop new tools to enhance drug delivery across BBB. Many signaling pathway that are active during BBB development are unraveled through mainly reverse genetic approach and gene knock out studies. Wnt/β-catenin signaling well known for its role in organ development, morphogenesis and in cancer causation is also reported to be essential for BBB induction and maintenance in most vertebrates. However many target molecules or effectors of this pathway remain to be identified. Previously we have identified Sox17 a SoxF family transcription factor, as a downstream molecule of Wnt/β-catenin signaling. It plays a key role in arterial differentiation of the vasculature of different organs. We found that Sox17 is also expressed at high levels in brain ECs throughout embryo development and in the adult vasculature. EC-specific gain of function of β-catenin (GOF) increases Sox17 expression in BBB ECs and may induce ectopic expression of Sox17 also in the choroid plexus vasculature that lacks the BBB. We hypothesized therefore that Sox17 might be involved in BBB development and maintenance downstream to the Wnt pathway. In Sox17 null mice we analyzed different functional characteristics of BBB such as permeability control and specific markers expression. The absence of this transcription factor leads to increase in BBB permeability to high molecular weight molecules and marked increase in PLVAP, a protein inversely related to maturation of BBB. We also observed significant reduction in the β-catenin signaling itself by employing BAT-gal reporter in Sox17 15 null background. In addition, many direct β-catenin targets like Axin2 and LEF1 were decreased upon Sox17 abrogation. These data suggested that Sox17 is not only a target of β-catenin signaling but also could maintain steady state, detectable levels of β-catenin signaling. We could rescue β-catenin signaling and correct BBB defects by either inhibiting the β-catenin destruction complex or by employing GOF β- catenin in Sox17 null background. Our study shows that Sox17 is an important regulator of BBB and it partially acts by sustaining β-catenin signaling. Sox17 expression and signaling may be important in pathological conditions like intracranial tumors and modulation of its activity could have clinical implications and therapeutic benefits.

Matteo Biancospino


Functional and structural characterization of myosin VI isoforms
Myosin VI is unique among the many members of the myosin superfamily. The peculiarities of this motor protein reside on its ability to travel along actin microfilaments towards their pointed-­‐ end, as well as its capacity to act either as anchor or as processive motor. As a consequence, myosin VI has been implicated in clathrin-­‐mediated endocytosis, vesicle trafficking, autophagy, cell migration and tumorigenesis. The current understanding of the myosin VI protein does not explain how it carries out these diverse processes, as functional mechanistic studies are lacking. Alternative splicing in the tail region generates myosin VI molecules with different features (myosin VIshort and myosin VIlong) but little is known about the impact of the variable region on physiological and pathological functions of myosin VI. In this study, we have analysed the myosin VI isoforms from a molecular and a functional perspective. Using quantitative mass spectrometry approach and modern NMR we identified and structurally characterized clathrin light chain as novel and isoform-­‐specific interactor. Within the novel clathrin-­‐binding domain that is unique to myosin VIlong, an isoform-­‐specific regulatory helix, named α2-­‐linker, defines a specific myosin VIlong conformation. Its presence or absence determines the target selectivity of myosin VIshort and myosin VIlong isoforms, acting like a molecular switch that regulates their functional involvement in migratory or endocytic pathways, respectively. The adaptor-­‐binding RRL motif is embedded in the clathrin-­‐binding domain and is masked in the structural configuration adopted by the myosin VIlong. Consequently, the previously identified RRL interactors show selective binding to myosin VIshort. Thus, we can provide for the first time a mechanistic explanation of why the various isoforms show different localization and function (i.e. myosin VIlong selectively involved in clathrin-­‐mediated endocytosis). We also found that alternative myosin VI splicing is aberrantly regulated in ovarian cancers, where exon skipping dictates myosin VIshort-­‐only expression. Importantly, cancer cell lines selectively expressing the myosin VIshort isoform exhibit severe migration defects when myosin VI is knocked down.

Maria Elena Bicchieri


A whole-genome approach to identify microRNA “modifiers” of breast cancer stem cell self-renewal
An emerging notion in the breast cancer field is that cancer stem cells (CSCs) are responsible for the degree of aggressiveness of the tumor and disease relapse. Therefore, the development of novel therapeutic strategies that specifically target the CSCs population could be the key to achieve an effective cure for breast cancer. Recent research has highlighted the role of microRNAs as key regulators of SC selfrenewal. The overall goal of this project was to identify key miRNA “modifiers” of breast cancer SC self-renewal with the purpose of identifying those involved in the regulation of stemcell traits and bona fide novel therapeutic targets. We used a lentiviral microRNA library to perform a functional whole-genome screening based on the self-renewal ability of CSCs on a breast cancer cell line. The screening yielded to 20 candidate microRNAs selected as potential modifiers of selfrenewal. A proof-of-principle validation revealed that 6 out of 10 tested cloned, confirmed their effects even when analyzed as single clones, underlining the potentiality of the whole-genome phenotype screening. We focused our attention on the two most promising candidates and, in order to search for the mechanism through which these microRNAs exert their function, we performed an RNA-seq analysis. We revealed that one microRNA in particular was able to control independently different pathway related to self-renewal, migration and proliferation, suggesting that this miRNA could effectively act at multiple levels to silence the selfrenewal potential of cancer stem cells and, likely, inhibit the proliferation and migratory ability of the tumor, too.

Heide-Marie Binder


Sleeping Beauty finds cooperating partners of oncogenic Myc
Myc triggers a transcriptional program inducing hyper-replication and proliferation but also tumor suppressive mechanisms like apoptosis. Therefore, Myc dependent tumors display high selective pressure to accumulate secondary mutations blocking these tumor-suppressive pathways. In the Eμ-myc mouse model, Myc is constitutively expressed in the B-cell linage under the control of the immunoglobulin heavy chain enhancer. The most prominent failsafe program known to be disrupted in Eμ-myc lymphomas is the p53-ARF pathway. In order to find novel cooperating partners of Myc leading to transformation we applied a conditional Sleeping Beauty transposon-based mutagenesis screen in vivo. By adoptive transplantation of Eμ-myc hematopoietic progenitors we generated 312 experimental animals, prone for lymphoma onset. Data show a strong genetic cooperation between the Eμ-myc transgene and SB transposon mobilization with accelerated tumor onset. Arising lymphomas were of the pre/pro- and immature B-cell stage and infiltration included extra-lymphatic tissue like the liver. The genomic sequences immediately adjacent to integrated transposons of 184 lymphomas were enriched by ligation-mediated PCR and were sequenced with a multiplexed approach. Based on published (Brett et al., 2011) and new bioinformatic methods, we identify 338 common integration sites (CIS) of which 188 were found mutated in human B-cell lymphoma. Pathway and GO term analysis reveal modulation of the Ras-MAP-kinase signaling pathway. Next to well-known modulators of Myc-induced lymphomagenesis including Bcl-XL, p53, ARF and Mdm2, we find CIS that were not yet reported in the Eμ-myc lymphoma like Map3K5.

Sara Bisi


Novel IRSp53 (Insulin Receptor Substrate protein of 53 kDa) functions in actin capping, filopodia formation and polarity establishment
IRSp53 (Insulin Receptor Substrate protein of 53 kDa) is a multidomain scaffolding protein that possesses an I-BAR domain (responsible for protein dimerization and membrane deformation), a partial CRIB motif (for the interaction with active CDC42), a SH3 domain (through it binds actin-related proteins) and additional protein interaction modules at the C-terminus. We have studied a previously unknown capping role of IRSp53 and the functional consequences of its interaction with VASP. In particular, we showed that following CDC42 binding and activation, IRSp53 is able to recruit and cluster VASP both in vitro and on the plasma membrane, thus fostering VASP processive elongation activity. The elongation of actin filaments by VASP, together with the membrane-deforming ability of IRSp53, combine to promote filopodia initiation downstream of CDC42. Thus, IRSp53 is essential for coordinating these activities in vitro and in cells in culture. In the second part of this thesis, an IRSp53 role in the polarity program was investigated. We showed that IRSp53 is apically localized in 3-D cyst models of epithelial cells, and IRSp53 depletion leads to a multilumen phenotype in MDCK and Caco2 spheroids. Moreover, using MDCK cysts as a model system, we noted that IRSp53 is enriched early at the prospective lumen initiation site (called AMIS), even before the apical marker podocalyxin. Finally, IRSp53 removal affects podocalyxin localization in MDCK cysts at the early stages. We speculate that IRSp53 might be involved in the trafficking of the apical protein podocalyxin, we are currently investigating the molecular mechanism responsible for this putative role.

Margherita Bodini


Genomics of treatment response in acute myeloid leukaemia
Acute Myeloid Leukemia (AML) is a cancer of the myeloid lineage of blood cells. Despite the high rates of complete remissions achieved after treatment (60-80% in young adults), the number of patients that will result cured after induction and consolidation therapy is low (~12%). The molecular basis of relapsing disease is still unclear and the few identified predictive factors has small predictive power. Currently, chemotherapy induction treatment consists in the administration of mainly three drugs (fludarabine, cytarabine, and idarubicin). In this thesis, endowed of the NGS technologies advancement, we decided to delineate the possible process of relapse formation in order to be able in the future to predict which patients are more susceptible to relapse. Our experimental plan includes the whole exome analysis of 30 pairs of primary/relapsed AML samples using NGS to identify relapse-specific mutations, the bioinformatics analysis of the clonal evolution of the disease and the identification of pathways that correlate with the relapsing disease. Our analysis shows that the genomic landscapes of primary and relapse AMLs are similar and in the majority of the patients (76%) some relapse clones were already present in the primary tumour and reappeared after chemotherapy at similar or augmented cellular frequencies. In 4 out of 29 patients (14%) we were able to identify driver mutations in the blood sample of the complete remission at low frequency; we hypothesize that more sophisticated diagnostic tools, based on NGS analysis, would help in driving the treatment to obtain better outcomes for patients.

Federico Boem


A matter of style. How map thinking and bio-ontologies shape contemporary molecular research
The aim of this thesis is to provide an epistemic analysis of the transformations occurring in contemporary biological research by considering the relation between molecular biology and computational biology. In particular, I will focus on bio-ontologies, as the tool which incarnates at best the new face of biomedical research. Such a choice is not arbitrary. By appealing to the notion of style of reasoning and way of knowing, I will show that bio-ontologies exemplify the rise and success of map thinking as the signature of a new way of doing molecular biology, while the theoretical tenets, established more than 30 years ago, still maintain their epistemic prominence. This is neither to say that experimentalism will disappear from science, nor that the experiments power will be diminished but rather that experiments will have a new role in the architecture of scientific efforts, precisely because of the increasing importance of classificatory approaches. Therefore, such a transition within biomedical research is indeed radical and profound but it does not involve paradigm shifts but rather a change in the practice. In this sense, it is a matter of style.

Maria Elena Boggio Merlo


NPMc+ as a model system to investigate the role of quiescence in leukemia development
The evolution of Acute Myeloid Leukemia (AML) is a complex process characterized by the stepwise accumulation of mutations, primarily occurring in Hematopoietic Stem Cells (HSCs). Such mutations give rise to the so-called Leukemia Initiating Cell (LIC), characterized by enhanced self-renewal and impaired differentiation. The molecular mechanisms underlying this transition are still poorly understood but they are likely to be critical to understanding the leukemic stem cell (LSC) biology. Recent functional and genetic studies on AML revealed NPMc+ as a critical driver oncogene, highly conserved at relapse, and characterizing the AML phenotype. Likely, NPMc+ has a pivotal role in the LIC evolution and LSC behavior. Taking advantage of the extended pre-leukemic phase of our inducible NPMc+ mouse model, we elucidated the impact of NPMc+ expression on normal HSCs to define the early mechanisms of NPMc+ induced leukemogenesis. We have found that NPMc+ expression leads to the expansion of the HSC compartment by enforcing a stem-cell transcriptional program that promotes quiescence and increases self-renewal. Moreover, considering the strong co-occurrence of NPMc+ with FLT3-ITD in patients, we investigated the mechanisms of this cooperation in pre-leukemia. Strikingly, the expression of NPMc+ in the FLT3-ITD background i) prevents the HSCs exhaustion imposed by FLT3-ITD, ii) restores their repopulating capacity, iii) restores the same transcriptional program observed in the NPMc+ HSCs, including quiescence genes upregulation. These data strongly suggest that NPMc+ and FLT3-ITD mutations cooperate in inducing AML, thanks to the NPMc+ ability to limit LT-HSCs exhaustion and reconstitute a fully competent LT-HSC population in which the oncogenic activities of FLT3-ITD lead to a rapid selection of the LICs. We thus hypothesized that enforced quiescence might be critical to maintain the transformed clone during both the pre-leukemic and the leukemic phase. In support, we xi identified the TGFβ pathway, one of the most critical pathways that regulate HSCs quiescence, as being upregulated by NPMc+, either alone or in combination with FLT3- ITD. Moreover, we report that pharmacological inhibition of this pathway impacts on NPMc+/FLT3-ITD AML growth in vivo. Finally, we provide preliminary results suggesting that TGFb inhibition might modify the fitness and/or the number of LSCs. xii

Paolo Bonaiuti


A fistful of molecules: cells escape an operational mitotic checkpoint through a stochastic process
The cell cycle culminates with the segregation of sister chromatids, which is a fundamental step in ensuring the transmission of unaltered genetic material. Chromosome segregation is carried out by the mitotic spindle, which captures and pulls sister chromatids towards the opposite poles. Anaphase starts when the correct bipolar attachment is achieved. Chromosomes migrate evenly to the two daughter cells, both inheriting the same genetic material. The presence of unattached kinetochore at anaphase onset is dangerous, since it may lead to unbalanced ploidy of daughter cells, with severe consequences for their survival. For this reason, improperly attached chromosomes activate the mitotic checkpoint that arrests cell division before anaphase. Cells can maintain an arrest for several hours but eventually will resume proliferation, a process we refer to as adaptation. Whether adapting cells bypass an active block or whether the block has to be removed to resume proliferation is not clear. Likewise, it is not known whether all cells of a genetically homogeneous population are equally capable to adapt. Here, we show that the mitotic checkpoint is operational when yeast cells adapt and that each cell has the same propensity to adapt. Our results are consistent with a model of the mitotic checkpoint where adaptation is driven by random fluctuations of APC/CCdc20 , the molecular species inhibited by the checkpoint. Our data provide a quantitative framework for understanding how cells overcome a constant stimulus that halts cell cycle progression.

Pranami Bora


An integrative approach to identify binding partners of Myc using (epi)genomics data in the 3T9MycER, Eμ-myc and tet-MYC
The c-MYC oncogene encodes the transcription factor Myc, which regulates a large number of biological processes and is overexpressed in a large number of cancers. When overexpressed, Myc binds to almost all open promoters but only regulates specific subsets of genes. We investigated this issue in three systems where Myc is overexpressed: 3T9MycER fibroblasts, Eμ-myc B cells and tet- MYC liver cells, through an approach integrating different types of next generation sequencing data, such as DNase-seq footprinting, ChIP-seq and RNA-seq, with motif analysis and machine learning methods (random forest). In order to analyse the DNase-seq footprinting data in our systems, we developed a novel pipeline that carries out step-by-step analysis of the raw DNase-seq data, and outputs DHS and TF footprints. We overlapped genome wide the footprints identified by the pipeline with matches of a PWM library, obtaining a list of footprinted PWMs. We first applied a single feature classifier assessing the performance of each of the PWMs one by one, and we found that single PWMs only provided a limited classification of the gene subsets. We then turned to a random forest classifier that considers combinations of PWMs as features. This strategy provided a good separation of the data sets (AUC>0.7) and identified some candidates, such as Nrf1/Nrf2 (Eμ-myc T up), Tead factors (Eμ-myc T and tet-MYC up), E2f4 (Eμ-myc T up) and E2f1(Eμ-myc T and tet-MYC up), that could potentially act with Myc in regulating specific subsets of genes.

Luca Bravi


β-­‐catenin signaling in CCM3 null endothelial cells contributes to Cerebral Cavernous Malformation pathology
Cerebral cavernous malformation (CCM) is a vascular disease that affects blood vessels in the central nervous system, which become malformed, leaky and prone to hemorrhage. The organ location is critical, both for neurological consequences and therapeutic intervention, which is exclusively surgical to date. The cause of CCM can be either sporadic or genetic. Mutations in three genes named CCM1/Krit1, CCM2/Malcavernin and CCM3/Pdcd10 are associated with hereditary CCM. Here we describe a novel murine model of the disease that develops after endothelial cell-selective ablation of the CCM3 gene in newborn mice and that we use to investigate the molecular mechanisms behind the development of CCM. We report enhanced transcription activity of b-catenin in CCM3-knockout endothelial cells in in vitro and in vivo models and we demonstrate that such activation is critical at early stage of the pathology development. In particular, we found that b-catenin is fundamental to trigger the process of endothelial-to-mesenchymal transition (EndMT), which has been previously demonstrated to be crucial for CCM development. Noteworthy, the activation of b-catenin pathway results Wnt-independent, while it correlates with the impaired state of endothelial cell-to-cell junctions typical of vessels developing CCM lesions. We also show that a pharmacological screening of a panel of drugs targeting b-catenin signaling revealed the NSAIDs sulindac sulfide and sulindac sulfone as potent inhibitors of this signaling pathway in endothelial cells. Moreover, we found that sulindac sulfide and sulindac sulfone are able to attenuate b-catenin transcription activity and to significantly reduce the number and dimension of vascular lesions in the central nervous system of mice with endothelial cell-specific CCM3-knockout. These NSAIDs thus represent pharmacological tools for inhibition of the formation of CCM3 vascular lesions, particularly with a view to patients affected by the genetic variant of CCM, who continue to develop new malformations over time.

Serena Buontempo


Functional dissection of Ezh2 during neuraldifferentiation
Epigenetic control of developmental genes is a key mechanism to achieve a time-regulated expression during differentiation as demonstrated to date in several differentiation pathways. In embryonic stem cells these developmental genes are marked by histone H3 lysine 4 (H3K4me3) and histone H3 lysine 27 (H3K27me3) trimethylation. The preservation of these bivalent marks is essential for proper time-regulated expression. It has been shown however that H3K27me3, catalyzed by Ezh2 within Polycomb Repressive Complex 2 is critical for the entire process of differentiation. Thus during neural differentiation, Polycomb targets have been shown to be dynamic with new genes acquiring or losing the mark at each step of neural differentiation. However most of the efforts were done on lineage commitment while Ezh2 role in the late phases of neural differentiation is less studied. In this work the role of the epigenetic axis focusing on Ezh2 loss in post-mitotic neurons was dissected. To this end, after ES cells derivation from Ezh2 conditional mice, a neural differentiation protocol was used to produce a homogeneous population of glutamatergic post-mitotic neurons. The lack of Ezh2 does not impair survival and development of neurons in culture. Although genome-wide transcriptomic approach revealed that Ezh2 is needed in post-mitotic neurons to maintain proper gene expression of genes involved in glutamate transmission, in particular our analysis shows an imbalance in glutamate receptors subunits expression. Thus our data suggest that post-mitotic neuronal maturation might be epigenetically controlled through Ezh2 regulation of glutamatergic pathway.

Claudia Burrello


Gut microbiota crosstalk with conventional and non-conventional T cells: a game of many players
The presence of microbial commensals in the gut requires the establishment of a complex network of reciprocal interactions between the microbiota and the host immune system to allow nutrient absorption while preventing undesired mucosal immune responses. Despite these homeostatic mechanisms, during intestinal inflammation alterations of the microbiota composition, namely dysbiosis, trigger abnormal immune responses. Here, we aimed at investigating the functional crosstalk between gut microbiota and the mucosal immune system during inflammation and upon induction of microbial dysbiosis. We observed that inflammation-induced and antibiotic-driven types of dysbiosis are phenotypically and functionally modifying CD4+ T and iNKT cells activity. Moreover, during intestinal inflammation, the experimental manipulation of the microbiota community through Faecal Microbiota Transplantation (FMT) reduces colonic inflammation and initiates the restoration of intestinal homeostasis through the induction of IL-10 production by immune cells. Further, we performed a comprehensive analysis on intestinal iNKT cells isolated from surgical specimens of active Inflammatory Bowel Disease (IBD) patients and non-IBD donors. We report that the exposure to mucosa-associated microbiota drives iNKT cell pro-inflammatory activation, inducing direct pathogenicity against the intestinal epithelium. Collectively, we provided solid evidence that a strict crosstalk between the gut microbiota and the intestinal conventional and non-conventional T cells exists. Antibiotic-associated dysbiosis has immunostimulatory functions. Moreover, FMT can therapeutically control intestinal experimental colitis and this poses FMT as a valuable therapeutic option in immune-related pathologies. In addition, we generated fundamental knowledge about the pathogenic functions exerted by human intestinal iNKT cells upon the interaction with mucosa-associated microbiota communities.

Valentina Buttiglione


The role of miR-340 in post-transcriptional regulation of the uPA-system in breast cancer
The urokinase-type plasminogen activator system (uPA-system), whose main components are the serine protease uPA (PLAU), the cell surface receptor uPAR (PLAUR) and the uPA inhibitor PAI-1 (SERPINE1), has been extensively studied for its involvement in cancer pathogenesis. Specifically, nowadays the components of the uPA-system are well-characterised determinants for the prognosis of breast cancer. The regulation of the gene expression of the uPA-system components is very complex and depends on a plethora of stimuli acting both at transcriptional and post-transcriptional level. The uPA-system components are often over expressed in breast cancer but the detailed molecular mechanisms regulating the expression are still to uncover. In an expression analysis conducted on a cohort of unselected breast cancer patients, we found that the expression of PLAU and PLAUR is highly correlated. Meta-analyses of published experimental data and in silico studies pointed out the possibility that PLAU, PLAUR and also SERPINE1 might be negatively regulated at post-transcriptional level by a microRNA, the miR-340. We experimentally validated the role of miR-340 as negative regulator of the expression of the three uPA-system components using MDAMB- 231, a triple negative breast cancer cell line. Microarray experiments, performed to characterise the global transcriptome changes induced by miR-340 in MDA-MB-231 cells, showed that miR- 340 down regulates also the expression of desmoplastic reaction-related genes underlining a possible role of miR-340 in regulating tumour-associated genes. Notably, most of the identified miR-340 target genes were found indeed to be associated with poor clinical outcome in breast cancer. Functional studies carried out in MDA-MB-231 cells suggested that miR-340 might modulate cell proliferation, even if this effect was not confirmed in vivo. In order to better define the functional role of miR-340, we generated a miR-340 deficient mouse model, taken advantage of the zinc finger nuclease technology. Overall these data identify, for the first time, a single microRNA that is able to down regulate the expression of the three main components of the uPAsystem together with desmoplastic reaction and breast cancer prognosis-related genes, thus representing a new potential player in the pathogenesis of breast cancer.

Manuel Carminati


Molecular implications of Afadin and Aurora-­‐A–mediated phosphorylation of NuMA in spindle orientation
Oriented cell divisions contribute to tissue morphogenesis and homeostasis. Planar divisions occurring with the spindle within the epithelial plane enlarge sheets and tubules, while asymmetric cell divisions with the spindle aligned to the apico-basal polarity axis sustain differentiation programs. Several pathways have been involved in establishing correct spindle orientation, both in cultured cells and in vivo. Most of these pathways impinge on the evolutionarily conserved Gαi/LGN/NuMA complexes that orient the spindle by generating pulling forces on astral microtubules (MTs), via direct interaction of NuMA with the MT-motors Dynein/Dynactin. My PhD project focused on the molecular mechanisms underlying the spindle orientation function of Afadin, and on the relevance of NuMA phosphorylation by Aurora-A for spindle orientation. During planar cell divisions, Gαi/LGN/NuMA assemblies are restricted to the lateral cortex, for molecular reasons that are still unclear. Studies conducted during this thesis indicate that LGN interacts directly with the junctional and F-actin binding protein Afadin, and defined the TPR domain of LGN (hereon LGNTPR) and a Cterminal peptide of Afadin (AfadinPEPT) as the minimal interacting regions retaining micromolar binding affinity. The crystal structure of the LGNTPR-AfadinPEPT fusion protein shows that the AfadinPEPT threads along the LGNTPR superhelix with opposite chain directionality, similarly to what observed for LGN in complex with other ligands, including NuMA. Consistently, we provided evidence that Afadin competes with NuMA for binding to LGN. Afadin knock-down in HeLa cells leads to reduced LGN cortical levels, and unexpectedly also to complete loss of cortical NuMA and Dynein/Dynactin, and hence spindle misorientation. Importantly, we discovered that Afadin interacts concomitantly with F-actin and LGN in vitro. Furthermore, we 2 showed that loss of Afadin impairs correct cystogenesis of Caco-2 cells, suggesting that it plays essential functions in epithelial planar cell divisions. Altogether our data suggest a model whereby in metaphase Afadin mediates cortical recruitment of Dynein/Dynactin, by targeting LGN at the lateral cortex via direct and concomitant interaction with LGN and with cortical F-actin. Later, LGN engages with NuMA and Dynein/Dynactin to exert pulling forces on the mitotic spindle. Thus, Afadin represents the first described mechanical anchor between the acto-myosin cell cortex and Dynein/Dynactin MT-motors. Besides being spatially regulated, the cortical recruitment of Gαi/LGN/NuMA is timely controlled by mitotic kinases coordinating spindle orientation with mitotic progression. It was reported that the activity of the mitotic kinase Aurora-A is required for correct spindle orientation in human cells in culture, and that NuMA is among its phosphorylation targets. However, whether NuMA is phosphorylated directly by Aurora-A and how molecularly its kinase activity affects spindle orientation was still unknown when we started our studies. Analyses in HeLa and RPE-1 cells revealed that in metaphase depletion or inhibition of Aurora-A leads to aberrant accumulation of NuMA at the spindle poles and loss from the cortex, despite LGN localizes normally at the cortex. FRAP experiments revealed that Aurora-A governs the dynamic exchange between the cytoplasmic and the spindle polelocalized pools of NuMA. Experiments in vitro and in cells showed that Aurora-A phosphorylates directly three serine residues on the C-terminus of NuMA, and mutation of Ser1969 into alanine recapitulates the aberrant polar accumulation of NuMA and the spindle orientation defects observed upon Aurora-A inhibition. Thus we concluded that phosphorylation on Ser1969 of NuMA by Aurora-A controls 3 NuMA distribution between the spindle poles and the overlying cortex, and allows proper spindle orientation. Intriguingly, Ser1969 lies within a previously characterized MT-binding domain. In vitro co-sedimentation and bundling assays revealed that the binding affinity of NuMA for MTs is unaltered by Aurora-Amediated phosphorylation, suggesting that unphosphorylated NuMA accumulates at spindle poles via a receptor other than MTs. Most interestingly, with our experiments we also identified a new MT-binding domain of NuMA positioned downstream of the LGN binding motif. This result implies that NuMA can simultaneously interact with LGN and MTs. Based on these findings, we propose that in metaphase the MTbinding activity of NuMA may contribute to anchor astral MT +TIPs at cortical sites together with LGN.

Alessandro Carugo


Dissection of the molecular pathways involved in pancreatic cancer initiation and progression with a novel in vivo approach
Abstract not available

Lucia Casoli


Myc-dependent transcriptional programs in mammary epithelial cells
The transcriptional programs governing the decision of mammary epithelial stem and progenitor cells to self-renew or differentiate are still not completely outlined. Previous evidences pointed out the role of Myc, and in particular of its repressive activity with Miz1, in these processes. Since the presence of stem-like cells within the tumor, the so-called cancer stem cells (CSCs), is now considered crucial for tumor initiation and maintenance, clarifying this aspect of Myc biology could be relevant in understanding its contribution to the genesis of breast cancer. Here, we used RNA-seq technology to profile the transcriptional programs regulated by Myc in two different settings. First, we studied immortalized mammary epithelial cells, in which we induced either Myc loss or gain of function. In this setting, Myc contributed to the positive and negative regulation of different sets of genes. Activated genes are involved in proliferation, metabolism, ribosomal biogenesis, mitochondrial organization, chromatin modification, RNA processing and modification. Repressed genes, on the other hand, were mainly involved in lysosome and vesicle-mediated transport, angiogenesis, cell death, extracellular matrix interaction, cell adhesion regulation, epithelial development and morphogenesis. Second, we studied the effect of Myc activation in mammosphere cultures, which provide a measure of stem cell activity. We demonstrate that Myc, when overexpressed, is able to promote self-renewal of mammary epithelial stem cells, as assessed by increased mammosphere expansion, and confirmed by mammary gland reconstitution assays in vivo. This activity of Myc is in part dependent on the interaction with the co-repressor protein Miz1, since the Myc mutant V394D (hereby Myc VD), impaired in Miz1 binding, is defective in promoting self-renewal. Overexpression of Myc in mammospheres was associated with the de-regulated expression of about three thousand genes, with similar numbers of up- and down-regulated genes. A group of around nine hundred genes was specifically repressed by Myc WT and not by the VD mutant. Surprisingly, the overlap between the groups of regulated genes in those mammary epithelial cells in adhesion or grown as mammospheres was limited, illustrating the context-dependency of Myc-dependent responses. Thus, transcriptional repression via Miz1 may constitute one of the mechanisms through which Myc sustains mammary epithelial cell selfrenewal. We are currently setting a functional screen, among genes repressed in a Miz1-dependent manner, to identify those that are critical in this process. Our study shall shed light on the mechanisms through which Myc regulates self-renewal in mammary epithelial stem and progenitor cells. Understanding this could be crucial in order to clarify the physiopathological roles of Myc in the mammary gland.

Laura Cedrone


The role of enhanced Polycomb Repressive Complex2 activity in tumorigenesis
Polycomb Group of proteins are essential factors present in cells’ nuclei. These multiprotein complexes are key repressive chromatin factors that regulate cellular differentiation during development, contributing to the correct establishment of lineage-specific transcriptional programs. Moreover, they represent key factors of proliferation and deregulation of their levels and activity have been linked to the onset and development of several human cancers. Recently, gain of function heterozygous EZH2 mutations have been discovered in non-Hodgkin lymphomas and melanomas. These mutations cause an aminoacidic substitution within the EZH2 catalytic SET domain (Y641), resulting in increased H3K27me3 deposition. Very little is known about this mutated enzyme, therefore the aim of my thesis is trying to unravel the tumorigenic mechanisms underlying these mutations. To understand a general oncogenic role for this mutated enzyme, we used MEF as an alternative, simpler model system. We observed increased deposition of H3K27me3 without any relevant transcriptional alteration at steady state, confirming our results also in lymphoma cell lines. To investigate a cooperative transcriptional deregulation for mutant EZH2, we then subjected MEFs to three different stimuli (starvation, myc upregulation and reprogramming to pluripotency). Since we found this to be true only during cell-fate transition, we proposed a model in which the levels of the H3K27me3 are increasingly deposited where the mark is already present at steady state. This could be relevant in lymphomas, impeding centroblasts differentiation and resulting in tumorigenesis in the presence of concomitant oncogenic mutations. This observation could shed light on the molecular mechanisms underlying lymphomagenesis in patients.

Claudia Cella


Development of biodegradable nanoparticles for targeting Tumor Associated Macrophages: synthesis, investigation of the role of the surfactant and surface decoration in complex media
Tumor Associated Macrophages (TAMs) are involved in cancer proliferation, thus strategies to deplete them represent promising tools for chemotherapy. Pharmacological agents with multiple activities such as curcumin and RNA interference have been proposed; however their employment in therapeutics has been limited because of low systemic bioavailability. Accordingly, this thesis described as an innovative therapeutic approach for cancer treatment the development of polymeric nanoparticles (NPs) able to (i) increase pharmacokinetics properties of biomacromolecules and poor water soluble drugs, and (ii) guarantee TAMs specific targeting. The safe and versatile polymer Poly(Lactic-co-Glycolic)Acid (PLGA) has been used for the synthesis of NPs by both single (OW) and double (WOW) emulsion-solvent evaporation techniques. Different synthetic parameters have been taken into consideration, with particular focus on the surfactant role. As alternative to the commonly used Poly Vinyl Alcohol (PVA), a newly synthetized polymer (amino-PVA) and Calcium Sterate (CSt) have been investigated for their ability to modulate surface charge and biocompatibility. NPs with solid or core-shell structures, whose size was tailored between 200 and 300 nm, were obtained and a thorough characterization has been performed, with the help of innovative techniques such as single particle optical extinction and scattering (SPES) method. Both amino-PVA and CSt stabilized NPs were found to be able to load curcumin and biomacromolecules, either alone or in combination. Strategies for surface decoration with the employment of D-mannose as specific molecule to guarantee TAMs recognition were proposed. Finally, cytocompatibility of the amino-PVA and CSt stabilized NPs have been assessed.

Aurora Cerutti


Oncogene-induced altered DNA replication dynamics
Oncogene Induced Senescence (OIS) is a tumor suppressive barrier that blocks cell cycle permanently. OIS results from a robust DNA damage response (DDR) activation due to oncogene-induced hyper-proliferation. By performing a whole genome analysis of DNA replication dynamics occurring upon oncogene activation, I discovered that oncogene activation alters DNA replication by increasing replication fork speed and fork stalling, while decreasing the frequency of replication initiation. This is accompanied by a prompt DDR activation. As cells approach senescence the frequency of initiation increases, the level of fork stalling and fork speed decreases. Oncogene activation leads to DNA replication stress mainly at fragile sites and since telomeres resemble fragile sites, I then demonstrated that oncogene activation impairs telomere replication, by increasing fork stalling at telomeres. This is accompanied by increased fragile telomeres, stochastic telomeric attrition and persistent telomeric DDR. These results revealed a novel link between oncogene activation and telomere dysfunction, refining the model underlying OIS establishment. Oncogene activation increases Reactive Oxygen Species (ROS). Beyond their toxicity, ROS are essential second messengers mediating mitogenic signalling. We discovered that oncogene-induced ROS is mediated by the NADPH oxidase NOX4. Upon oncogene activation, NOX4 pharmacological inhibition blocks ROS production, resulting in fork speed reduction and differential regulation of local replication origin initiation. These results revealed a fundamental role of NOX4 and ROS in mediating oncogeneinduced hyperproliferation. Polycomb repressive complexes (PRCs) repress genes involved in development, proliferation and differentiation by EZH2-mediated H3K27 trimethylation. Recent independent studies revealed a more direct PRCs role on S phase progression and DNA replication. We show that EZH2 KO leads to impairment of cell cycle progression, with cells blocked at G1/S transition. Furthermore EZH2 KO impairs DNA replication, by reducing fork speed, increasing the frequency of initiation and fork stalling, demonstrating that PRCs deficiency leads to replication stress.

Luca Chiapperino


From consent to choice: the ethics of empowerment-based reforms
The aim of my thesis is twofold. First, I focus on the controversies arising from the renegotiations of patienthood and citizenship entailed in what I call ‘empowerment-based reforms’ (EBRs). What I define as EBRs will have in fact different implications for the various stakeholders involved in their development and implementation. Empowered citizens within EBRs will have access to (and will be required to manage) an unprecedented amount of information regarding their health conditions. Factors such as genetic and biological makeup, life-style behaviours and environmental exposures will be increasingly used (by both citizens and professionals) to identify treatment options, to target developing diseases, and to adopt preventive measures for future illnesses. Among the effects that this personalising vision of healthcare is likely to foster, it is thus worth emphasizing how the nature and scope of individual responsibility for health will be affected by this paradigm shift, and how this future scenario can be made an ethically desirable one. In my thesis, I therefore identify the range of normative exercises entailed in EBRs, and I present a normative analysis of empowerment aiming at highlighting the distinctive ethical aspects of this approach. Second, the goal of my thesis is to explore how the normative theorization of empowerment proposed above can accommodate one of the most pressing societal implications of epigenomics. Namely, its burdening of individual responsibility for health. On the one hand, I argue that novel approaches to prevention, diagnosis and treatment brought about by epigenomics are a fundamental tenet of the personalization project at the basis of what I call EBRs. In this respect, epigenome-based healthcare is thus likely to foster controversies similar to other epistemic endeavours of personalized medicine (e.g. genomics, metabolomics, pharmacogenomics), which can be addressed from the normative premises of empowerment. On the other, I maintain that concerns arising from the translation of epigenomics into healthcare practice should be poised with its promise to make increasingly visible the ‘contextual nature’ of health (i.e. tracing the mechanistic interaction between lifestyle, living conditions and individual health). Rather than limiting societal appraisal of epigenomics to the danger of burdening individual responsibility for health, I argue that epigenetic knowledge may become pivotal in fleshing out social and environmental influences inherently affecting individual health. Sufficiently valid, reliable and actionable epigenetic knowledge may in fact orient individual choice across the spectrum of environmental and lifestyle exposures determining health, thus championing epigenomics with the potential of serving the empowering aims fleshed out throughout this work. The road connecting the constitution of an empowered citizenship in healthcare, and the societal appraisal of epigenomics can be regarded as a two-way road. There is in fact a possibility that empowerment and epigenomics may respectively shape their normative and epistemic dimensions in the future of healthcare. It is thus towards the identification of the possible challenges and opportunities that this synergy may bring about that the theoretical attention of this work is devoted.

Ramveer Choudhary


Mechanisms controlling the integrity of converging forks during replication termination
During S phase,natural fork pausing elements including replication termination zones (TERs) and transcribed genes, can easily lead to!genotoxicity!and!chromosome!fragility at fragile sites (CFs), known as hotspots for DNA breaks and chromosomal rearrangements, particularly in cancer cells. However, the key factors and pathways protecting the integrity of CFSs are not well understood. In order to elucidate these mechanisms, we used yeast as a model system, and combined genetic and genomic studies aimed at identifying fragile sites genome wide. We found that centromeres, rDNA, telomeres and TERs represent hot spots for copy number variation (CNV) in wild type yeast cells exposed to replication stress. We found that top2 contributes to relieve the topological stress and extensive fork pausing during Termination at TERs. We also observed that fragility at these sites is greatly enhanced in checkpoint mutants (Tel1ATM1 and Mec1 ATR1), likely due to their inability to uncouple transcription from gene gating, thus leading to topological stress in front of replication forks. TER fragility in checkpoint defective! cells can be partially rescued by Top2 and Condensin complex inactivation. We further demonstrated that Pif1 family helicase, Rrm3, regulate extensive fork pausing at TERs and endogenous pausing sites by suppressing RNA DNA hybrid (RJ oops) accumulation. Alongside, a RNA DNA helicase Senataxin1 (Sen1) coordinates replication and transcription collision at TERs and endogenous fork pausing. Absence of Sen1 accumulates CNVs at TERs and double strand DNA breaks. We show thatprogrammed fork pausing and resolution of R loops are key processes for TER integrity. Accordingly, sen1%rrm3 double mutants, accumulate gaps and RNA DNA hybrids at TERsand fail to fuse replicons; moreover, these mutants exhibit unscheduled condensationevents at TERs leading to chromosome entangling. Taken together, our data strongly suggests that, the replication checkpoint, Sen1 and Rrm3, coordinate replication termination to prevent accumulation of unsolved topological constrains and premature recruitment of Condensins andTop2.

Cecilia Claudi


Polo-like kinase Cdc5 contributes to mitotic spindle elongation via the kinesin-5 motor protein Cin8
Proper chromosome segregation requires an orderly sequence of events, whereby spindle elongation follows the dissolution of sister chromatid linkages. The mitotic spindle is a sophisticated and complex machinery built of microtubules, microtubule associated proteins and motor proteins. Proper spindle function requires that microtubule dynamics are stabilized at anaphase. This change in microtubule dynamics is dictated by a shift in the balance of kinase and phosphatase activities in favor of the phosphatases. The finding that cells simultaneously lacking the polo-like kinase Cdc5 and the phosphatase Cdc14 cannot progress through anaphase albeit having cleaved cohesin due to defects in spindle elongation, challenges the view of mitotic exit as a time for protein dephosphorylation. We identified the kinesin 5 motor protein Cin8 as a key target of the “Cdc14-Cdc5” spindle elongation pathway. We show that besides being dephosphorylated by Cdc14, Cin8 is also phosphorylated by Cdc5 on residues S409 and S441, and that this phosphorylation is crucial for the function of the kinesin in anaphase spindle elongation. The finding that Cin8 is simultaneously a substrate of a kinase and a phosphatase sheds light on the complexity of mitotic exit regulation. Since it appears that phosphorylation and dephosphorylation events are equally important to the point that kinases and phosphatases cooperate to regulate the same substrates, the view of mitotic exit as the realm of phosphatases is dismantled and the continuous need for single molecule studies in addition to global analyses investigation is put forward.

Agnese Collino


Characterization of genetic and epigenetic modifications in a model of inflammation-driven cancer
Chronic inflammation is causally associated to many types of tumor, and has been recently acknowledged as a cancer hallmark. Nevertheless, whether inflammation has an intrinsic mutagenic potential is still not directly proven or understood from a mechanistic point of view. Furthermore, it is as yet unclear whether inflammation could induce epigenetic modifications, and if these changes are relevant to tumor generation. Therefore, the aim of this work was to assess inflammation-derived genomic and epigenomic modifications at multiple stages of tumorigenesis in Mdr2-knockout mice, a model of purely inflammatory hepatocellular carcinoma (HCC). By ChIPseq profiling of H3K27Ac mark we reported the establishment of an inflammatory program specifically in hepatocytes starting from the pre-malignant, chronic inflammatory step. This inflammatory signature is retained up to the more advanced HCC stage, and is accompanied by the activation of members of the AP1 transcription factor family. In parallel, by whole exome sequencing, we observed a high frequency of copy number amplifications and a very low number of point mutations in HCC nodules. Copy number variations occurrence was directly correlated to the grade of malignancy in each lesion. The JNK pathway was shown to be pervasively targeted by gene amplification, and to be involved in the adenoma-to-carcinoma transition. A comparable genetic landscape has been observed in a human liver cancer with similar etiology. In conclusion, this study shows that in a model of inflammation-driven cancer an epigenetic inflammatory signature is early acquired and maintained throughout disease progression. On the contrary, genetic alterations appear only at later stages and mainly target the JNK pathway. Future dataset integration will help clarifying chronological relationship and possible mutual interplay between mutations and epigenetic changes.

Arianna Colosio


Mechanisms mediating replication fork collapse and processing in checkpoint defective cells
An accurate DNA replication is essential to prevent genome instability events, such as mutations and chromosomal rearrangements that are hallmarks of neoplastic transformation and cancer onset. A dedicated branch of the DNA damage checkpoint maintains the integrity of replicating chromosomes by stabilizing replication forks in the presence of genotoxic agents, thus ensuring cell viability. Upon fork collapse, budding yeast checkpoint mutants experiencing replication stress accumulate aberrant replication intermediates, such as gapped and hemi replicated molecules, as well as four-branched structures known as reversed forks. Aberrant replication intermediates are potentially harmful for the cells since they are thought to trigger unscheduled recombination events that cause genome rearrangements. In this PhD thesis, I examined checkpoint-dependent mechanisms controlling fork stability, and I provide in vivo evidence that positive supercoiling accumulating ahead of replication forks is the main mechanical force driving fork reversal. Thus, DNA topology is a critical determinant of replication fork stability in vivo. Furthermore, a 2D-gel screening for enzymatic activities involved in the metabolism of collapsed forks, revealed a novel role for the Sae2 and Dna2 endonucelases in replication intermediates processing.

Alexia Conte


Integration of modeling and experiments to define principles of EGFR activation and ubiquitination under physiological and pathological conditions
Epidermal growth factor receptor (EGFR)-dependent signaling is involved in numerous physiological processes, and its deregulation leads to cellular dysfunctions and pathologies, first and forecast, cancer. Endocytosis has a crucial impact on the downstream EGFR signaling response and it is regulated by ligand concentration. Indeed, depending on the EGF dose, the EGFR can be internalized through clathrinmediated endocytosis (CME) or non-clathrin endocytosis (NCE). The switch between these two internalization mechanisms occurs over a narrow range of EGF concentrations (1-10 ng/ml). Importantly, EGFR ubiquitination shows a threshold response over the same range of EGF doses and is responsible for the commitment of EGFR to NCE, and thus, for EGFR signal extinction through receptor degradation. In this project, we were interested in elucidating the cellular mechanisms that regulate and coordinate the choice between these two endocytic routes, in addition, we aim to clarify how the integration of the two pathways influences EGFR downstream signaling. In order to deal with the complexity of the system, we adopted an integrated research approach combining mathematical modeling with wet-lab experiments. To this purpose, in collaboration with the Systems Biology group at our Institute, we developed a mathematical model of early EGFR activation that quantitatively accounts for the ubiquitination threshold observed at 2 minutes of EGF stimulation. The ‘early model’ was able to generate important predictions; in particular, it predicts a weakness in the system that is unveiled in the presence of high EGF concentrations and EGFR overexpression, two conditions frequently observed in cancer. We tested these predictions using different cell-based model systems subjected to varying perturbations. A challenge in the biological validation of the model, was obtaining quantitative reproducible data. To this aim, we optimized a quantitative ELISA-based assay to measure EGFR ubiquitination/phosphorylation upon different perturbations. This assay revealed to be powerful and allowed us to validate the predictions generated by the model. Thanks to our integrative approach, we identified Cbl as the limiting and weak element of the system. We expect that our model of EGFR activation will provide novel insights into the role of EGFR endocytosis, controlling the balance between EGFR signaling and downmodulation, frequently altered in cancer.

Andrea Corno


Decoding Molecular and Phenotypic Properties of Proliferating Cells under constant Mitotic Checkpoint Activation
During mitosis, DNA material needs to be properly segregated. Chromosome segregation is triggered by the Anaphase Promoting Complex or Cyclosome (APC/C), an E3 ubiquitin ligase activated by its cofactor Cdc20. Activation of APCCdc20 is conditioned by the presence of optimal conditions for a proper partition of sister chromatids. Without such conditions, activation of APC/CCdc20 could lead to unequal chromosome segregation, by which an aneuploid progeny could arise. The Spindle Assembly Checkpoint (SAC) inhibits the activity of APC/CCdc20, blocking the progressionthrough the cell cycle and thus preventing erroneous chromosome segregation. When cells experience a prolonged SAC activation, they may die in mitosis by apoptosis, or overcome the arrest and progress into the cell cycle even when chromosome segregation is impaired. The second scenario is also known as adaptation to the SAC (or mitotic slippage). Once adapted, cell proliferation can still be blocked (via apoptosis or G1 arrest) or cells can resume cell division. The latter case may establish a progeny of cells in which aneuploidy and genomic instability introduce large genetic variability, with potentially irreversible and deleterious effect on the cell population. Using S. cerevisiae as a model organism, we characterized a population of cells escaped from a prolonged mitotic arrest, which we called adapted cells. Proteomic analysis of these cells revealed large rewirings of biological processes and pathways, suggesting a pseudo "differentiated" state for adapted cells. The cell cycle of adapted cell is heavily modified, to account for the chronic inhibition of APC/CCdc20. On the one hand, APC/CCdc20 itself become less responsive to the SAC, as observed in a population of cells where we uncoupled adaptation from missegregation. We showed that cellular size was not responsible for the partial recovery of APC/CCdc20 activity in the presence of the SAC. Our data rather suggest a role for Cdc28-mediated phosphorylation. On the other hand, other activators of APC/C like Cdh1 become essential, unlike what observed in a regular cell cycle. The synthetic lethality of adapted cells with mitotic exit genes suggests potential molecular targets for specific inhibition of adapted cells.

Ottavio Croci


Genomic landscape and transcriptional regulation by YAP and Myc in the liver
This thesis is divided in three sections; the main project and two appendixes. In the main project we studied YAP, the downstream effector of the Hippo pathway, a transcriptional co-factor that plays a fundamental role in dedifferentiation, cell proliferation and transformation. While its upstream regulation has been extensively studied, its role as transcriptional co-factor is still poorly understood. We show that YAP co-adjuvates the transcriptional responses of the Myc oncogene to promote cell proliferation and transformation. In addition, we show that YAP promotes cell de-differentiation by antagonizing liver-specific transcriptional programs controlled by transcription factor HNF4A, thus providing a mechanism on how YAP can promote de-differentiation of hepatocytes. In the first appendix we investigate the mechanism of BRD4 inhibition, a promising strategy for the treatment of Myc-driven tumors. We demonstrate that the efficacy of this strategy relies on the control of transcriptional elongation mediated by BRD4 on gene promoters, independently of the downregulation of Myc oncogene. Although the inhibition of BRD4 causes its genome-wide displacement from promoters of virtually all expressed genes, the effects on transcription are restricted to a subset of sensitive genes. Our results show how the impairment of elongation genome-wide can affect specific transcriptional programs. In the second appendix we describe a new web application, Chrokit, aimed at analyzing genomic data in a fast and intuitive way. The application is multiplatform and can be run on dedicated servers to maximize computational power and provide accessibility to multiple users simultaneously.

Alessia Curina


Dissection of the mechanisms controlling high constitutive activity of housekeeping and tissue-specific cis-regulatory elements
The genetic information is identical within the organism but the mechanisms by which different cell types achieve specialized functions interpreting the same set of instructions is not completely understood. It is now increasingly accepted that the combination of different genomic elements, both promoters and enhancers, favors the recruitment of different TFs, which in turn promotes the assembly of different preinitiation complexes, guaranteeing heterogeneity in transcriptional outputs across different tissues. Nevertheless, the cis-regulatory elements and the transcriptional rules that control and maintain the expression of constitutively active genes are still poorly characterized. Specifically, whether the constitutive activity of promoters and enhancers relies on entirely distinct or instead shared regulators is unknown. By dissecting the cis-regulatory repertoire of macrophages, we found that the ELF subfamily of ETS proteins selectively bound within 60 bp from the transcription start sites of highly active housekeeping genes. ELFs also bound constitutively active, but not poised macrophage-specific enhancers and promoters. The role of ELFs in promoting constitutive transcription is suggested by multiple evidences: ELF sites enabled transcriptional activation by endogenous and minimal synthetic promoters; ELF recruitment was stabilized by the transcriptional machinery, and ELF proteins mediated recruitment of transcriptional and chromatin regulators to core promoters. These data indicate that a distinct subfamily of ETS proteins imparts high transcriptional activity to a broad range of housekeeping and tissue-specific cisregulatory elements, which is consistent with the role of an ETS family ancestor in core promoter regulation in a lower eukaryote.

Roberto Cuttano


KLF4 is a key determinant in the development and progression of Cerebral Cavernous Malformations
Cerebral cavernous malformations (CCMs) are capillary-venous malformations located in the central nervous system often resulting in cerebral hemorrhage. Pharmacological treatment is needed, since current therapy is limited to neurosurgery. CCM affects up to 0.5% of the human population and occurs in both sporadic and familial forms. Loss-of-function mutations in any of three genes CCM1, CCM2 and CCM3 have been associated to familial CCM. Postnatal endothelialspecific deletion of any of the three Ccm genes in mice results in the development of brain vascular malformations that faithfully resemble human CCM lesions. Here we describe that CCM are formed by endothelial cells (ECs) undergoing endothelial-to-mesenchymal transition (EndMT). Ccm1 deletion leads to activation of the MEKK3-MEK5-ERK5-MEF2 signaling cascade resulting in a marked upregulation of the transcription factor Krüppel-like factor 4 (KLF4) in ECs in vivo. KLF4 promotes an endogenous production of bone morphogenetic protein 6 (BMP6) in ECs that, in turn, activates the transforming growth factor-β (TGF-β) and BMP signalling pathway. KLF4 transcriptional activity and KLF4-dependent TGF-β/BMP pathway activation are responsible for the EndMT switch observed in the absence of Ccm1. Interestingly, using both a pharmacological treatment to inhibit TGF-β/BMP pathway or a genetic approach based on endothelial-specific Ccm1 and Klf4 double knockout mice, we strongly reduce the development and progression of CCM lesions. Importantly loss of Klf4 almost abolishes mouse mortality in endothelial Ccm1-ablated mice. These data indicate that KLF4, TGF-β/BMP pathway and EndMT are crucial events for CCM pathogenesis and unveil KLF4 as a key therapeutic target for CCM.

Giuseppe D'Agostino


Propagation of dysregulation across gene expression layers in 7q11.23 CNVassociated developmental disorders
Williams-Beuren Syndrome (WBS) and 7q11.23 microduplication syndrome (7dup), two multisystemic developmental disorders, arise from symmetrical copy number variations of the same region on chromosome 7q. In WBS patients this region is deleted, whereas it is duplicated in 7dup individuals. These syndromes display a striking combination of shared and opposite clinical manifestations at the level of neuro-cognitive, craniofacial and cardiovascular features, thus pointing to a remarkable dosage-sensitive effect of a small group of genes on the development and maintenance of complex traits such as sociality, language and facial morphology. We derived a large cohort of induced pluripotent stem cells (iPSCs) from samples with WBS, 7dup and from healthy controls, and we demonstrated that, already at the pluripotent state, the transcriptome is dysregulated in pathways that map onto disease-related features. Moreover, these pathways were selectively dysregulated in differentiated lineages, thus demonstrating an anticipatory power of the pluripotent state. Building on these results, we expanded the view on the dysregulation in pluripotency by measuring three layers of gene expression: transcriptome, translatome and proteome. We mapped the propagation of differences across layers by integrating ribosome profiling and SWATH-MS proteomics, and we probed the extent to which a translation initiation factor included in the CNV, EIF4H, was responsible for the regulation of translation. We found that each layer of gene expression has its own differentially expressed genes, whose degrees of propagation can change between layers. Moreover, differentially expressed genes can cluster by different ways of propagation when they are compared to the levels of EIF4H.

Carolina D'Alesio


Identification of novel epigenetic targets that sustain breast cancer growth
Breast cancer is the second leading cause of tumor-related death in women, mainly due to resistance to first line therapy, high risk of relapse and metastatic dissemination. Breast cancer is a highly heterogeneous disease, which displays diverse biological characteristics, clinical behaviour and prognosis. For these reasons, it has become challenging the identification and characterization of novel genes responsible for breast cancer initiation and progression. To identify new targets that sustain breast cancer growth, we performed in vivo and in vitro shRNA screens in a human breast cancer cell model. We screened two libraries targeting several chromatin remodeling enzymes (around 200 in total), which are essential genes in breast cancer maintenance and represent optimal druggable candidates. We identified approximately 70 genes that were depleted in our screens, and among them, we selected five hits to validate the screens. Remarkably, the silencing of each target gene significantly reduced tumor growth in vivo and decreased proliferation, colony formation and migration in vitro, thus validating our screens. We deeply investigated the Chromodomain Helicase DNA binding Domain 4 (CHD4) gene, whose silencing in breast cancer cells greatly reduces tumor growth, but does not affect normal mammary epithelial proliferation and migration. We examined the role of CHD4 in primary cells derived from spontaneous mammary tumors of MMTV/NeuT transgenic mice. Upon CHD4 depletion, we confirmed a significant decrease of tumor growth in vivo and cell proliferation and migration in vitro. Intriguingly, we demonstrated that CHD4 silencing reduced tumor growth in vivo in a patient-derived xenopatient (PDX) model of Luminal B drug-resistant breast carcinoma. Moreover, we investigated the mechanism through which CHD4 promotes breast cancer cell proliferation and we showed that CHD4 regulates cell cycle progression of breast cancer cells. CHD4 depletion provokes a consistent accumulation of cells in the G0/G1 phase and a strong reduction of the S phase of the cell cycle, and an upregulation of p21. In summary, RNAi screens allowed us to identify CHD4 as a critical target that sustains human breast cancer. Importantly, we showed that CHD4 modulation does not modify normal mammary cell proliferation and migration, suggesting that its targeting in tumor cells might not impact on the surrounding normal tissues. Moreover, CHD4 is crucial in almost any subtype of breast cancer, as shown by its effect on MMTV/NeuT and PDX tumorigenesis. Finally, we demonstrated that CHD4 is a key regulator of breast cancer cell cycle.

Veronica D'Uva


Relevance of NUMB isoforms in breast cancer
Numb acts as tumor suppressor in the human mammary gland by inhibiting Notch signaling and stabilizing p53. Loss of Numb is detected in ~30% of breast cancers and correlates with an adverse prognosis. The best characterized mechanism for Numb loss is degradation of the protein by the ubiquitin-proteasome machinery. At the molecular level, Numb binds to and inhibits the activity of Mdm2, the E3 ligase that mediates p53 ubiquitination and successive degradation. A short Numb region encompassing the alternatively spliced exon 3 is necessary and sufficient to inhibit Mdm2 and stabilize p53. The aim of this project is to define the function of Numb isoforms in p53 regulation and their clinical relevance in breast cancer. We demonstrated the exclusive role of Numb isoforms containing the exon 3-coded insert in regulating the p53 circuitry. Alterations in the Numb splicing pattern have clinical implications in breast cancer. Indeed, p53 WT tumors displaying reduced levels of exon 3-containing Numb isoforms showed increased chemoresistance and risk of relapse. A similar effect was also observed in tumors displaying low levels of Numb mRNA, revealing that transcriptional deregulation represents an additional mechanism responsible for loss of Numb expression. Our data demonstrate that specific Numb isoforms and global Numb mRNA levels correlate with chemoresistance and tumor progression in p53-competent breast cancers.

Maria Damjanovicova


Epigenetics and Policy: cross-linking the ‘environmental turn’ in the life sciences and the ‘molecular turn’ in epidemiology
This thesis is an empirical investigation of epigenetics and policy. This research first focused on mapping the impact of epigenetics in health care and exploring the challenges it poses for health care policy. The thesis developes a combined qualitative-quantitative strategy to identify the most active areas of epigenetic research, clinical applications and clinical outputs, as well as to track the number of publications on epigenetics. Moreover, the thesis finds that the science of epigenomics goes ‘beyond the genome’ insofar as what lies beyond can be conceptualised through and converted into genome-friendly, code-compatible digital representations. The research further focused on the case of Glasgow, a city characterized by stark health and social inequalities, where epigenetics has been employed in an interdisciplinary project to measure and instruct relevant social programs to target these inequalities. This thesis thus contributes a critical insight into how epigenetics is currently employed – in collaboration between actors of diverse backgrounds; and in policy efforts and action upon health. The thesis finds that within this project epigenetics is conceptualised as instrumentally effective, policy-approprate evidentiary resource that could foster socio-political change in a non-distant future. Accordingly, it is thanks to its molecularization of the environment and therefore its purported objectivity, that epigenetics is bestowed the potential for actionable public health knowledge. Additionally, the thesis finds that it is the solidary practice that governs this interdisciplinary collaborative endeavour in Glasgow.

Vivek Das


Leveraging transcriptomic analysis to identify transcription factors orchestrating cancer progression
Next generation sequencing (NGS) technology is currently employed to explore the molecular profiles associated to different biological contexts. The application of this technology provides at same time a high-resolution and global view of the genome and epigenome phenomena, enabling us to study the molecular events underlying many human diseases, including cancer.  Our lab tries to exploit the utility of high throughput sequencing technologies generating genomic, transcriptomic and epigenomic data from patient’s cohort to study the underlying molecular mechanisms that characterize the specific diseases and map the key regulators that can be critical targets for relevant therapeutic measures. I take the advantage of this technology to mainly understand two aggressive cancers: Ovarian Cancer (OC) and Glioblastoma multiforme (GBM). OC is a leading cause of cancer-related death for which no significant therapeutic progress has been made in the last decades. Also, in this case, despite multimodal treatment its prognosis remains extremely poor. This is due to the fact that the molecular mechanisms underlying OC tumorigenesis and progression are still poorly understood (Vaughan et al., 2011). GBM is the most common and aggressive primary brain malignancy with very poor prognosis (Frattini et al., 2013). The median survival rate is of 12-15 months (Singh et al., 2012) with 5-year survival that is less than 5% despite the multimodal treatment which include  surgery, radiotherapy and chemotherapy. To this end, I will be integrating various genomic and transcriptomic analysis to define the key regulatory actors that characterize the disease progression paving. This integrated analysis has been devised in form of a computational workflow that gives way for a discovery pipeline for physiopathologically meaningful epigenetic targets that can lead to therapies.

Giulia De Conti


in vivo shRNA screening to identify quiescence-related genes required for AML growth
AML is hierarchically organized with at the apex Leukemia Stem Cells (LSCs), a rare cell population able to initiate and sustain the tumor growth. LSCs share many functional properties with normal Hematopoietic Stem Cells (HSCs) including self-renewal capacity and quiescence. Quiescent LSCs can survive to radiation and chemotherapy acting as a reservoir for leukemia relapse, the major cause of death for AML patients. Therefore, LSCs quiescence is critical for leukemia maintenance and few evidences suggest that quiescence regulation in pre-leukemic phase plays a pivotal role for leukemogenic process as well. In this work, we demonstrated that the expression of NPMc+ or PML-RARα in HSCs is sufficient to enforce a quiescent stem cell gene expression profile. Therefore, we hypothesized that enhancement of the quiescent phenotype in HSCs could be a shared mechanism for leukemia development and maintenance. As an approach to examine the contribution of representative quiescence related genes in AML, we exploited RNA interference technology to perform in vivo screening. Among the target genes we found depleted during the screening, silencing of Stat1 or Sytl4 in AML blasts was sufficient to significantly decrease in vitro self-renewal and delay leukemia growth in vivo.

Valentina De Lorenzi


Cross-talk between the proteolytic and non-proteolytic functions of the urokinase receptor
Urokinase (uPA) and its cellular receptor (uPAR) have been implicated in many pathological events, such as tumour cell migration and dissemination. The binding of uPA to uPAR favours extracellular proteolysis by enhancing plasminogen activation. Moreover, it promotes cell adhesion and signalling through binding to the provisional matrix protein vitronectin (VN). We here report that plasminogen activation induces a negative feedback on cell adhesion to VN, mainly through the proteolytic cleavage of the RGD-motif in VN. The cell-adhesive properties of VN are impaired by disruption of the integrin binding site and release of the somatomedin B (SMB) domain responsible for uPAR binding. VN represents the first described uPAR-dependent substrate of uPA. Our findings therefore identify a potential novel function of uPAR in focusing the activity of the plasminogen activation system onto extracellular matrix-associated VN. SMB-containing VN fragments are released by several cancer cell lines and detectable in human urines. Our hypothesis is that they might represent a novel cancer biomarker as a functional measurement of the uPA-system activity in the tumour tissue and we have developed a clinical grade immunoassay for the quantification of such fragments in urine samples. Finally, we show that the urokinase inhibitor, PAI-1, blocks the feedback mediated by uPA and behaves as a uPA-dependent agonist of the uPAR-VN interaction. We report that the uPA·PAI-1 complex displays higher agonistic activity than uPA. These data might represent a molecular explanation for the poor clinical outcome observed in cancer patients with high levels of uPA and PAI-1.

Lorenzo Del Savio


Private Choices, Public Issues. The Ethics of Health Policy in the Face of Diet Related Diseases.
The epidemic of diet related disease is a fundamental fact of epidemiology that our societies are increasingly facing. It calls for policy responses and amendments of our systems of health care. These actions and reforms intersect several loci of moral and political disagreement in the public sphere: the acceptability of public paternalism, the appropriate consideration of personal responsibility in health care and the moral and political significance of social health inequalities. I offer a treatment of these three broad normative issues in order to inform discussions about appropriate responses to diet related diseases. (1) I argue that antipaternalism is overstated if not understood in welfares terms: within the latter framework, evidence for poor capability in dietary choices is a sound reason for intervention. (2) I distinguish distributive and efficiency concerns regarding personal responsibility for health, arguing that there is no defensible conception of the former. (3) I dismiss efforts to understand the moral importance of social health inequalities in terms of health entitlements and reject investment-like approaches to inequalities framed in terms of "equality of opportunity": the fight against health inequalities is vivified by a renewed interest in the social goods attached to robust socio-economic egalitarianism. Together, these three theses lead away from policies focused on individuals, their responsibility and their productive importance for society and support both public health interventions on the environment where people live and continuous defense of traditional unconditional health care provision.

Maira Di Tano


Fasting-mimicking diet-based non-toxic combination therapy in cancer treatment: from molecules to bedside
Fasting and fasting-mimicking diet (FMD) were shown to delay tumor progression and sensitize a wide range of tumor types to the toxic effect of chemotherapy, while protecting normal cells, through a mechanism that can involve the lowering of IGF-1 and glucose levels. In my thesis, I have focused my studies on the identification of non-toxic, but effective interventions for cancer treatment. For this purpose, I evaluated the effect of FMD on the highly aggressive KRAS mutant colorectal cancer, which is one of the most widespread and lethal cancer in the western world. Throughout my studies, I identified a FMD-based non-toxic intervention, in which FMD cycles are combined with a non-toxic compound, for the treatment of KRAS mutant cancers. I found that the FMD-based combo therapy is able to selectively kill KRAS mutant tumor cells, while leaving unaffected KRAS wild type cancer cells and normal cells. Moreover, the combinative treatment results to be safe and effective also in in vivo mouse models, reducing tumor progression in both immunocompromised and immunocompetent mice. I also provide evidence for the mechanism explaining the synergistic effect of FMD and the selected non-toxic compound. In addition, my results indicate that this combo therapy enhances chemotherapy efficacy in KRAS-driven CRC mouse models, thus representing a promising therapeutic option, which can be hopefully translated into the clinic.

Elisa Donato


Mechanistic investigation of BRD4 inhibition in MYC dependent tumors
The c-myc gene encodes for a transcription factor involved in the regulation of different cellular mechanisms, ranging from cell cycle control to cellular metabolism. Myc is frequently altered in human cancer either by genomic rearrangement or by alteration of upstream regulatory pathways. Myc crucial role both in tumor formation and maintenance makes it an attractive molecular target for cancer therapy. Unfortunately, Myc is intrinsically resilient to direct pharmacological targeting. To overcome this issue, alternative therapeutic avenues have been explored. Independent groups showed that BET proteins inhibition leads to a strong Myc downregulation in Multiple Myelomas and Acute Myeloid Leukemias, with consequent cell cycle arrest and tumor regression. Depending on c-myc location (translocated versus endogenous), two different working models were proposed to explain BETi efficacy. In order to improve our understanding of the mechanism of action of BETi, we evaluated global transcriptional alteration and chromatin profiles in Burkitt’s Lymphomas in response to JQ1. Our results demonstrate that BETs inhibitors cause global alteration of RNA PolII dynamics, due to the role of BRD4 in regulating elongation. Nonetheless, the selective transcriptional effects following JQ1 treatment is due to promoter saturation and high RNA PolII pausing that render the expression of JQ1 sensitive genes rate-limited by transcriptional elongation. Indeed the same genes are selectively targeted by pharmacological treatments affecting components of the elongation machinery. These observations highlight the role of BETs protein in regulating gene expression and provide a rationale to explain how broad inhibition of elongation may lead to a selective transcriptional response.

Ambra Dondi


Adaptation to the DNA damage checkpoint requires the rewiring of the cell cycle machinery
The DNA damage checkpoint is a surveillance mechanism evolved to preserve genome integrity in response to DNA damaging agents. The DNA damage checkpoint senses DNA insults and halts the cell cycle providing time and conditions to repair the lesion(s). If the damage is successfully repaired, cells reenter in the cell cycle in a process known as checkpoint recovery. If the damage is not repaired, cells either die or override the checkpoint reentering the cell cycle in the presence of the lesion. This process, known as checkpoint adaptation, represents an opportunity for cells to repair the damage in the following cell cycle. However, checkpoint adaptation can be an unsafe event as daughter cells can accumulate genomic aberrations. Indeed, checkpoint adaptation has been described to occur also in cancer cells. The players involved in the adaptation process remains largely unknown. In budding yeast S. cerevisiae, it was shown that the activity of the Cdc fourteen early anaphase release (FEAR) network, a pathway that control the exit from mitosis, is required for checkpoint adaptation, therefore suggesting a crosstalk between the cell cycle machinery and the DNA damage checkpoint. We found that the activity of single FEAR components is dispensable both in unperturbed conditions and following repair of the DNA lesion (i.e. checkpoint recovery). However, the activity of single FEAR components is strictly required for exit from mitosis in persistent DNA damage conditions, therefore suggesting checkpoint adaptation as the rewiring of a cell cycle with peculiar features.

Benedetta Erba


Abnormalities of the endothelial lineage in primary myelofibrosis
Primary Myelofibrosis (PMF) belongs to the family of Myeloproliferative neoplasms (MPNs), a heterogeneous group of related clonal malignant diseases characterized by the oncogenic transformation of the multipotent hematopoietic progenitor cell (HPC), which leads PMF patients to develop massive bone marrow (BM) fibrosis. Clinical hallmarks also include progressive splenomegaly, anemia and weakness due to ineffective hematopoiesis and excessive production of pro-inflammatory cytokines, which play a central role in mediating increased deposition of BM stromal fibers. So far, no curative treatment for this pathology exists with the exception of allogenic stem cell transplant, thus highlighting the great need to find an alternative and less risky therapy. Although fibroblasts are directly implicated in fibrosis development, endothelial cells (ECs) can also play a role in PMF; indeed, when activated by inflammatory cytokines such as TGF-b, they undergo a process called Endothelial-to-mesenchymal transition (EndMT), leading ECs to acquire fibroblastic features. Our results show that ECs can undergo EndMT during the development of PMF both in patients and in a MPN mouse model of the disease. This process occurs during early stages of fibrotic degeneration and is primarily mediated by the release of TGF-b by megakaryocytes (MKs) and platelets. Moreover, patients with different genetic mutations inducing PMF all undergo EndMT, thus proposing it as a common mechanism of fibrosis development. We also demonstrate that TGF-b induces endogenous BMP4 and BMP6 up-regulation in splenic ECs, further sustaining EndMT phenotype. Moreover, our results show that the in vitro treatment of splenic ECs with TGF-b and BMP inhibitors can revert EndMT phenotype, thus opening the possibility to the use of specific and more targeted therapy for PMF patients to achieve fibrosis remission.

Valentina Fajner


Characterization of the E3 ligase dHecw, a novel member of the Drosophila melanogaster Nedd4 family
Ubiquitination is one of the most abundant and versatile post-translation modifications in eukaryotes, and it plays an important role in many biological processes by affecting protein activity, interactions, localization and stability. E3 ligases (E3s) have a key function as ubiquitin-substrate matchmakers, providing specificity to the reaction, yet little is known about the targets and functions of the majority of E3s. We identified and characterized dHecw, a novel member of the Drosophila melanogaster Nedd4 family of ubiquitin E3s. dHecw is the single ortholog of the human HECT ligases HECW1 and HECW2, which are the less characterized members of the family. dHecw expression is tightly regulated in the central nervous system and in the ovary, and is down-modulated during aging. Catalytic inactive dHecw mutants and dHecw KO flies, generated by CRISPR/Cas9 technology, presented signs of neurodegeneration, as short lifespan, limited motor function and brain tissue vacuolization. They also showed premature decline in fertility due to germline specific defects in oogenesis. dHecw interactome, identified by mass spectrometry analysis, presents several ribonucleoparticles (RNPs) components, including dFmr1, a translational repressor that controls localized mRNA translation in developing embryo. We demonstrated that dFmrp is a dHecw substrate in vitro, and we found a genetic interaction among the two proteins. Our investigation of the functional outcome of dFmrp ubiquitination suggests that it does not cause dFmrp degradation but, instead, it impacts on its function/interaction network. Altogether, our data suggest that dHecw is a novel player involved in the dynamic regulation of RNPs required for neuronal health and fertility.

Paolo Falvo


Role of obesity in the development of acute promyelocytic leukemia (APL)
Obesity is a pathological condition characterized by an augmented presence of fat mass in the body, and a known risk factor for many cancer types. Clinical data have shown that the incidence of acute promyelocytic leukaemia (APL) is strongly correlated with obesity. APL is characterized by reciprocal translocation between chromosomes 15 and 17 resulting in the formation of the PML/RARα oncoprotein. However, the molecular mechanisms explaining the effects of obesity on APL development have not been elucidated. To recapitulate clinical observations, we characterized a mouse model of diet-induced obesity using transgenic mice constitutively expressing PML/RARα in the hematopoietic system (PML/RARα KI mice) and wild type mice as a control. Mice were fed standard diet (SD) or high fat diet (HFD), and leukaemia-free survival was monitored. We observed that HFD-fed PML/RARα KI mice developed leukaemia earlier and with higher penetrance than SD-fed mice. We evaluated the extent of DNA damage in hematopoietic stem cells (HSC) after four months of diet. We demonstrated that HFD-fed PML/RARα KI mice presented a 40% increase of DNA damage in HSCs as compared to SD-fed PML/RARα mice. However, this was not associated with an increased mutational load as revealed by whole genome sequencing-based method. We investigated whether HFD confers a proliferative advantage to PML/RARα bone marrow using colony forming assays. Indeed, we observed that HFD PML/RARα bone marrow has a higher clonogenicity than bone marrow from SD-fed mice. Finally, we demonstrated that linoleic acid (LA), which is the main component of HFD increases PML/RARα enhanced self-renewal.

Giulia Ferretti


Ethical issues in mammography screening programs and Women’s decision making
The dissertation deals with the ethics of mammography screening programs and related individual women’s decision-making regarding whether to participate the program. While, on the one hand, mammography screening is proven beneficial in reducing breast cancer mortality at the population level, on the other hand routine screening also entails potential harmful outcomes for individuals such as false positive results and overdiagnosis. Intrinsic difficulties in evaluating the magnitude of benefits and harms of mammography, as well as in the identification of the target population, gave rise to a tremendous scientific, political and social debate. The latest up-to-date Western countries’ guidelines recognize the complexity of assessing the net benefit of routine mammography screening, especially for women aged 40 to 49 who, nevertheless are eligible to participate programs on the basis of their personal evaluation and preferences. This dissertation argues that shared decision-making constitutes the most effective way of respecting women’s autonomy and incorporating their values in the decision-making process. Among different ethical theories, the dissertation claims that care ethics represents the most suitable normative ethical approach to justify and provide a concrete guide to SDM, which can also be applied to the inquiry of older women’s mammography screening as well as to other clinical contexts. Finally, in order to provide physicians with a practical tool for eliciting women’s values and preferences, the major themes of women experiencing breast cancer identified through a critical interpretative literature review are systematically presented and proposed as a basis for an empirical qualitative research on the issue.  

Giulia Fragola


Functional dissection of polycomb complexes in somatic cell reprogramming
Somatic cells can revert to a pluripotent state following ectopic overexpression of the four transcription factors (TF): c-Myc, Kl4, Oct4 and Sox2. Upon TF-induced reprogramming, cells undergo genome-wide resetting of both transcriptional and epigenetic states. Among the different epigenetic marks, Histone H3 lysine 27 tri-methylation (H3K27me3) has been proposed to play a major role in this process. Here I provide a functional dissection of H3K27me3 function in TF-induced reprogramming. To this purpose, I derived induced pluripotent stem cells (iPSCs) from mouse embryonic fibroblasts (MEFs) expressing a catalytically dead form of EZH2, the major mammalian H3K27 methyltransferase. Remarkably, loss of EZH2 activity did not affect reprogramming efficiency and was compatible with the derivation of iPSCs globally devoid of H3K27me3. Ezh2-deficient iPSCs were indistinguishable from Ezh2 control iPSCs at both molecular and functional level since they could efficiently silence the somatic transcriptome and differentiate into tissues derived from the three germ layers. Interestingly, genome-wide analysis of H3K27me3 in Ezh2 mutant iPSCs revealed the retention of residual levels of this mark on a highly selected group of Polycomb targets enriched for developmental regulators controlling the expression of lineage specific genes. Retention of H3K27me3 mark was guaranteed by the presence of an alternative PRC2 complex that compensated EZH2 loss by the recruitment of its homologue EZH1. Erasure of residual H3K27me3 from these targets led to a striking impairment in TF-induced reprogramming indicating that PRC2-mediated H3K27 trimethylation is required on a highly selective core of Polycomb targets whose repression enables TF-dependent cell reprogramming.

Marco Fumasoni


Replication-associated repriming and primer processing facilitate error-free DNA damage tolerance and sister chromatid synapsis
DNA damage tolerance (DDT) mechanisms are crucial for genome integrity as they allow efficient bypass of endogenously or exogenously generated lesions. Error-free bypass of lesions is accomplished by a recombination-related mechanism, generally referred to as template switching (TS), that allows the recovery of the damaged information from the sister chromatid. Pioneer studies revealed key enzymatic functions required for error-free DDT and identified sister chromatid junctions (SCJs) as crucial DNA intermediates mediating this process. However, little is known on the temporal window and the chromatin/topological context in which TS takes place. Using S. cerevisiae as a model system, here I investigated the contribution of different replication-related processes to recombination-mediated DDT. We identified sister chromatid cohesion and replication-related repriming/primer processing as novel pathways implicated in TS. Unexpectedly, repriming during replication impinges on sister chromatid cohesion, but these two processes differentially contribute to errorfree DDT. Repriming activities and other processes influencing the size of the ssDNA gap and the availability of the 5 end positively influence error-free DDT both in spontaneous and genotoxic stress conditions. Our results provide evidence that error free DDT is largely a recombination-mediated gap-filling process with different requirements from the recombination mechanism involved in double strand break repair, both in what regards the influence of DNA end processing on the subsequent homology search and strand invasion, and the contribution of the chromatin structural context mediated by sister chromatid cohesion. Our findings suggest that mechanisms promoting damage-bypass via replication-associated SCJ formation affect DDT pathway choice and the establishment of post replicative sister chromatid synapses.

Ilaria Galasso


Precision medicine in society: promises, expectations and concerns around social and health equity
This thesis analyzes precision medicine from an ethical and political perspective, especially in terms of distributive justice: it is aimed at investigating the kinds of benefits that can be produced after precision medicine, and the possible distributions of those benefits, by considering the consequent impact of precision medicine on social and health equity. Precision medicine is considered as a social construct subjected to different interpretations, and it is analyzed by mainly referring to two major case studies: the Precision Medicine Initiative in the US, and the 100,000 Genomes Project in the UK. The analysis focuses on the promises of precision medicine, as expressed in the discourses of the two projects, compared with the expectations and the concerns, as expressed in published comments and in fieldwork interviews with relevant stakeholders. The analysis investigated the scope of precision medicine with respect to public health, the inclusiveness of precision medicine, and the democratizing capacities. It emerged that there are different versions of precision medicine, which encompass different scopes and possibly produce different kinds of benefits. In particular, one version, by also including in its scope the social determinants of health, is argued to have the ‘societal potential’ to inform socio-political interventions to promote social equity and, in return, health equity. It is argued that, although the benefits directly deriving from precision medicine - tailored biomedical treatments and information supposed to empower individuals - risk to totally exclude socio-economically disadvantaged groups, thus preventing any solidarity-based participation, on the other hand, the implementation of the ‘social potential’ would foster the public good and a solidarity-based medicine, to the advantage of everybody. Some challenges for the actualization of this ‘societal potential’ are identified and discussed. The aim of this thesis is to contribute to overcoming those challenges by promoting the dialogue and the alignment between medical innovation and socio-political reforms.

Alessandro Galbiati


New approaches to study DNA double-strand breaks genome-wide and in single-cells
Cells have evolved several DNA repair mechanisms to maintain their genetic information unaltered and a DNA damage response pathway (DDR) coordinates DNA repair with several cellular events including a cell-cycle arrest until damaged DNA is repaired. When cells fail to repair DNA lesions, they undergo either apoptosis or cellular senescence. Tools commonly used to detect DNA lesions rely on indirect, antibody-based recognition of proteins associated to DNA breaks. Unfortunately, these tools do not allow direct and precise localization of the breaks, leaving several biological questions unanswered. I validated and optimized BLESS and BLISS, two methods that allow genome-wide single-nucleotide resolution mapping of DNA double strand-breaks (DSBs). Using these techniques, I studied the impact of DSBs on transcription. I characterized a DDR-dependent transcription inhibition around breaks. Differently, I observed that following macrophages LPS-stimulation, a transient wave of DSBs is induced at LPS-specific enhancers and it correlates with their transcription activation, thus suggesting a new mechanism for transcription activation involving controlled DNA damage generation.    BLESS and BLISS cannot be applied to single-cell studies. Thus, I developed a new method, named DI-PLA, for the detection and imaging of DSBs in fixed cells and tissues. I applied DI-PLA to demonstrate that senescence cells and aged tissues accumulate DSBs, which are associated with persistent DDR activation, that is known to fuel cellular senescence. Finally, I describe a modification of BLESS to discriminate between DSB bearing telomeres and deprotected telomeres which could be applied to further characterize the mechanisms of DDR activation at telomeres.

Sara Gallini


Molecular contribution of the Aurora-A kinase and the junctional protein Afadin to oriented cell divisions
Spindle positioning is essential for tissue morphogenesis and homeostasis. The orientation of the mitotic spindle is determined by force generators, Dynein/Dynactin, formed on cortical NuMA:LGN:Gαi complexes, which anchor astral microtubules at specialized domains of the cortex. In this context, my PhD project aimed to study the molecular mechanisms accounting for the spindle orientation functions of the Aurora-A kinase and the junctional protein Afadin. To understand the implication of Aurora-A in spindle orientation, I partially inhibited its activity. Under these conditions, in metaphase NuMA accumulates abnormally at the spindle poles without reaching the cell cortex. FRAP experiments revealed that Aurora-A governs the dynamic exchange between the cytoplasmic and the spindle-pole-localized pools of NuMA. At molecular level, Aurora-A phosphorylates directly the C-terminus of NuMA on three serine residues, among which Ser-1969 is the determinant for the dynamic behaviour of NuMA at the spindle poles. Collectively, my studies demonstrate that Aurora-A activity on NuMA controls its cell localization, and that this is the major event underlying the spindle orientation functions of Aurora-A in cultured cells. Part of my PhD studies addressed the role of Afadin in spindle orientation; I demonstrated that Afadin is required for spindle positioning, and epithelial morphogenesis of Caco-2 cysts. Molecularly, Afadin binds directly and concomitantly to F-actin and to LGN. Indeed, in mitotic HeLa cells, Afadin is required for cortical accumulation of LGN, NuMA and Dynein above the spindle poles, in a F-actin dependent manner. These results depicted Afadin as the first mechanical anchor between Dynein and cortical F-actin.

Pierre Luc Germain


Humans, animals, and Petri dishes: Biomedical modeling between experimentation and representation
The evaluation of biomedical models  in vitro and in vivo models of diseases  raises several methodological issues, and has generally been hampered by a lack of attention to the precise functions played by biomedical models. For if biomedical models are ultimately expected to inform us about human pathologies, they seldom do so in isolation, and get there through a wide variety of ways. An epistemological understanding of this process is therefore a precondition for their evaluation, and this thesis is an attempt at building such an epistemology. Several of the examples used come from cancer research, especially xenograft models and models used in the context of large-scale drug screenings. Another important set of examples come in vitro models, with a particular focus on disease modeling using induced pluripotent stem cells. I argue that the notion of model, if conceived as to apply to biomedical models, conflates into that of experimental system. I therefore propose an account of biomedical models that does not presuppose a fundamental divide between modeling and experimentation. I show that biomedical models are not simply scaled-down versions of their target, but instead projections of their target in a different space of representation. I argue for an instrumental role of biomedical models, and use this role to explore the diversity of proximal functions fulfilled by biomedical models. I propose the notion of distributed modeling to draw attention to the relations between model systems, and illustrate this by analyzing the interplay between in vitro and in vivo models. Finally, I explore the implications of this account for the evaluation of biomedical models, and more broadly for the topic of scientific representation.

Smbat Gevorgyan


Novel Poly(amidoamine) Nanoparticles Designed for Drug Delivery to the Central Nervous System
With the aging population, central nervous system diseases are becoming increasingly widespread. However, the treatment of such diseases represents a challenge mainly due to the presence of the blood-brain barrier, which effectively blocks most of drugs to pass into the central nervous system and impedes the treatment of diseases. Therefore, effective ways of drug transport to the central nervous system are highly required. Polymeric nanoparticles are drug delivery vectors that have high specificity to their targeted sites and increasingly enhance the stability of encapsulated drugs. Poly(amidoamines) are a promising family of synthetic polymers that can be used in nanoparticles synthesis. Nanoparticles synthesized from linear poly(amidoamines) show very good biocompatibility levels and specific targeting properties. Nevertheless, those nanoparticles are mainly synthesized by self-assembly and usually lack long-term stability. In this study we developed innovative poly(amidoamine) nanoparticles synthesized by ultraviolet light assisted photo-crosslinking. The synthesis method had an advantage of being simple, did not require toxic organic solvents and was shown to be easily scaled-up. Moreover, poly(amidoamine) nanoparticles were coated by Polysorbate 80 surfactant, which not only increased the nanoparticles stability but also gave them central nervous system targeting properties. Poly(amidoamine) nanoparticles were able to successfully encapsulate model therapeutic compounds and release them in a slow and controlled manner. Furthermore, the nanoparticles showed high permeability levels across an in vitro blood brain barrier model. The developed poly(amidoamine) nanoparticles show great potential to be used as delivery vectors to the central nervous system.

Francesco Ghini


The plasticity of miRNA pool: a novel approach to reveal mechanisms behind miRNA turnover
MicroRNAs (miRNAs) are a small (18-25nt long), evolutionary conserved, class of noncoding RNAs that appears as a major regulatory component of gene expression, implicated in virtually all known physiological and pathological processes. They act at post-transcriptional level by silencing the expression of a multitude of target mRNAs through various mechanisms, including target degradation and protein synthesis inhibition. As a result, the regulation of the miRNA pool is one of the critical events in the definition of cell identity and behavior both in physiology and disease. To date, the dynamics of miRNA degradation and the mechanisms involved in remain largely obscure, in particular, in higher organisms. We developed a pulse-chase approach based on metabolic RNA labeling to calculate miRNA decay rates at genome-wide scale in mammalian cells. Our analysis revealed heterogeneous miRNA half-lives, with many species behaving as stable molecules (T1/2 > 24 h), while others, including passenger miRNAs and a number (25/129) of guide miRNAs, are quickly turned over (T1/2 = 4–14 h). Decay rates were coupled with other features, including genomic organization, tran- scription rates, structural heterogeneity (isomiRs), and target abundance, measured through quantitative experimental approaches. This comprehensive analysis highlighted functional mechanisms that mediate miRNA degradation, as well as the importance of decay dynamics in the regulation of the miRNA pool under both steady-state conditions and during cell transitions.

Giovanni Giangreco


Characterization of Epsin3 function in the acquisition of a partial EMT state in breast cancer through E-Cadherin endocytosis
 

Seyed Amir Hosseini


Dissecting the role of lysine-specific demethylase1 (LSD1): identification of markers/effectors of sensitivity to LSD1 inhibitors in cancer
LSD1 is a flavin-containing amine oxidase that, by reducing the cofactor FAD, demethylates H3K4me1/2 and H3K9me1/2 at target loci in a context-dependent manner. LSD1 could be an attractive target for cancer therapy because of its deregulation in a number of cancers, including lung, breast, melanoma and hematological malignancies. Given the unsatisfactory clinical outcome associated with standard chemotherapy in acute myeloid leukemia (AML) and melanoma treatment, there is an essential need for new targets. Recently LSD1 have gained great interest for their use as anticancer therapeutics. However, the efficacy of LSD1 inhibitors is limited to a substantial subset of cancer cells. Thus, identification of good predictive biomarkers for sensitivity to treatment with LSD1 inhibitors will be of great value in determining the most suitable therapeutic setting. In this study we demonstrated that, LSD1 drives unrestricted cycling of cancer cells by directing repressing CDKN1A (p21) gene, which allows unrestricted G1-S transition. Inhibition of LSD1 suppresses G1 to S phase transition and cell proliferation in a p21-dependent manner and P21 provoked by LSD1 inhibitor could serves as a biomarker to verify pharmacological activity and a prognostic tool reflecting responsiveness to LSD1 inhibitors. Moreover, loss of p21 enables progression of cell cycle and rescues the LSD1 inhibitor phenotypes. Finally, we found that forced cell cycle inhibition either with p21 induction by HDAC inhibitors or directly by CDK inhibitors

Sri Ganesh Jammula


ChIP_QC, computational platform for multivariate epigenetic studies and its application in uncovering role of polycomb dependent methylations states
During my PhD tenure, I have been involved in developing a user-friendly crossplatform system capable of analyzing epigenomic data and further use it in understanding the role of the Polycomb Repressive Complex 2 (PRC2) in genome regulation. From current trending in epigenetics research, we can sense increasing ease of high throughput sequencing and greater interest towards genome wide epigenomic studies. As a result of which we experience an exponential flooding of epigenetic related data such as Chromatin immunoprecipitation followed by sequencing (ChIPseq), and RNA sequencing (RNA-seq) in public domain. This creates an opportunity for crowd sourcing and exploring data outside the boundaries of specific query centered studies. Such data has to undergo standard primary analysis, which with the aid of multiple programs has been stabilized courtesy to the scientific community. Further downstream, out of many, genome wide comparative, correlative and quantitative studies have proven to be critical and helpful in deciphering key biological features. For such studies we lack platforms, which can be capable of handling, analyzing and linking multiple interdisciplinary (ChIP-seq/RNA-seq) datasets with efficient analytical methods. With this aim we developed ChIP_QC, a user-friendly standalone computational program with an ability to support numerous datasets with high/moderate sequencing depth for performing genome wide analysis. First, using ENCODE project (Consortium, 2012) data, we illustrated few applications of the program by posing different biological scenarios and showed the comfort with which some known observations can be verified and also how it can be helpful in deducing some other novel observations. 17 Second, we were interested in understanding the functionality of the products generated through catalytic activity of PRC2. It is known that Lysine 27 of histone H3 (H3K27) undergoes posttranslational modification (PTM) and methylation is one such dominant PTM. Methylation on H3K27 can be either mono/di/tri-methylation form. Out of all three forms, it is very well demonstrated that trimethylation of H3K27 (H3K27me3) is PRC2 dependent and at the same time its role in gene repression is well characterized, but functional roles of other forms of methylation on H3K27 are still poorly characterized. For understanding this, we used mouse embryonic stem cells (mESC) as model system of our study and we were able to provide an extensive characterization of other forms of methylation, highlighting their differential deposition along the genome, their fundamental role in transcriptional regulation, and their indispensability during differentiation program. Using ChIP_QC and with other computational methods along with experimental evidences, our data demonstrated that the monomethylation of Lys27 (H3K27me1) is required for correct transcription of genes and positively correlates with trimethylated Lys36 (H3K36me3); on the other hand dimethylated Lys27 (H3K27me2), that we identified to be the principal activity of PRC2, prevents firing of non cell type specific enhancers.

Arun K. Kolinjivadi


Dissecting the role of BRCA2, Rad51 and SMARCAL1 in vertebrate chromosomal DNA replication
DNA replication is a fundamental macromolecular event that is essential for cell division. During each cell cycle the entire genome has to be precisely duplicated to ensure genome integrity and stability. In the process of DNA replication, replication forks encounter endogenous and exogenous lesions and these lesions have to be rectified to transfer stable genetic material to daughter cells. Emerging evidences connected a role for Homologous Recombination (HR) proteins to replication fork protection during unperturbed DNA replication. Since HR protein, BRCA2 and Rad51 are essential for cell survival we used Xenopus laevis cell-free extract to dissect the function of BRCA2 and Rad51 during chromosomal DNA replication. Using Electron Microscopy (EM) we show that BRCA2 and Rad51 function together to prevent single-stranded DNA (ssDNA) accumulation at and behind the forks during unperturbed DNA replication. We further discovered that BRCA2 mediates interaction between Rad51, Polymerase alpha (α) and Polymerase delta (δ) and this interaction likely promote efficient re-priming and polymerising activity at stalled replication forks to prevent ssDNA accumulation. Moreover we show that inhibition of replicative polymerases in the absence of Rad51 results in increased frequency of replication fork reversal activity. We further found that replication fork reversal is predominantly induced by an annealing helicase SMARCAL1 in the absence of Rad51. Collectively our findings indicate that, to prevent ssDNA accumulation and aberrant replication fork architecture, timely re-priming of Polymerase alpha mediated by Rad51 is essential. Hence, loss of Rad51 impact replication fork architecture, eventually resulting in chromosomal abnormalities.

Kamal Kishore


Development of Computational Tools to Study the Patterning of DNA and RNA Methylation in Healthy and Disease States
Epigenetics can be defined as the set of sequence independent processes that produces heritable changes in cellular information. These chromatinbased events such as covalent modification of DNA and histone tails are laid down by the co-ordinated action of chromatin modifying enzymes, thus altering the organisation of chromatin and its accessibility to the transcriptional machinery. Our understanding of epigenetic intricacies has considerably increased over the last decade owing to rapid development of genomic and proteomic technologies. This has resulted in huge surge in the generation of epigenomics data. Integrative analysis of these epigenomics datasets provides holistic view on the interplay of various epigenetic components and possible aberration in patterns in specific biological or disease states. Although, there are numerous computational tools available catering individually to each epigenomic datatype, a comprehensive computational framework for integrated exploratory analysis of these datasets was missing. We developed a suite of R packages methylPipe and compEpiTools that can efficiently handle whole genome base-resolution DNA methylation datasets and effortlessly integrate them with other epigenomics data. We applied these methods to the study of epigenomics landscape in Bcell lymphoma identifying a putative set of tumor suppressor genes. Moreover, we also applied these methods to explore possible associations between m6A RNA methylation, epigenetic marks and regulatory proteins.

Francis Kobia


Targeting Notch trafficking in human cancer cells
Notch signaling is prominently involved in cell fate decision and growth regulation in metazoan tissues. Because of this, Notch is often upregulated in cancer and current efforts point to developing drugs that block its activation. Notch receptor endocytosis towards acidic compartments is a recently appreciated determinant of signaling activation. The Vacuolar H+ ATPase (V-ATPase) is responsible for acidification of endocytic organelles and recently it has been shown that mutants in V-ATPase subunit encoding genes in model organisms display loss of Notch signaling phenotypes. In the first part of my graduate studies, we aimed at discovering whether pharmacologic reduction of V-ATPase activity affected Notch signaling. We found that administration of BafilomycinA1 (BafA1), a highly specific V-ATPase inhibitor decreases Notch signaling during Drosophila and Zebrafish development, and in human cells in culture. In normal breast cells, we have found that BafA1 treatment leads to accumulation of Notch in the endo-lysosomal system, and reduces its processing and signaling activity. In Notch-addicted breast cancer cells, BafA1 treatment reduces growth in cells expressing membrane tethered forms of Notch, while sparing cells expressing cytoplasmic forms. In contrast, V-ATPase inhibition reduces growth of leukemia cells, without affecting Notch activating cleavage. However, consistent with the emerging roles of V-ATPase in controlling multiple signaling pathways, in these cells Akt activation is reduced, as it is also the case in BafA1-treated breast cancer cells. Our data support V-ATPase inhibition as a novel therapeutic approach to counteract tumor growth sustained by signaling pathways regulated at the endo-lysosomal level. The functions of Notch throughout the life of an individual are varied and complex. This complexity is not sufficiently accounted for by the limited core of known Notch signaling components and a growing body of evidence attributes it to additional factors that determine whether, when and how Notch functions within a given context. Considering this, in the second part of my graduate work, we sought to identify novel genes that might influence Notch. Thus, we performed a high content immunofluorescence-based RNA interference screen of a pharmacologically-relevant subset of the human genome. To this end, we monitored how knockdown of specific genes perturbs the localization of the Notch-1 receptor in human breast cells under resting and signaling conditions. Here we present the screen setup, the primary screen results and the candidate follow-up strategy.

Veronica Krenn


Molecular mechanisms controlling the recruitment of the checkpoint protein Bub1 to kinetochores
Bub1 is an essential component of the spindle assembly checkpoint (SAC), a safety mechanism required for accurate chromosome segregation. Kinetochores are protein assemblies built on the centromeres and are essential for SAC activity. During my PhD, I investigated the molecular mechanisms that control the recruitment of Bub1 to kinetochores, a key step for correct SAC functioning. Using immunoprecipitation and localization studies, I discovered that in human cells Bub1 uses two of its subdomains to interact with the kinetochore component Knl1. One, the Bub3-binding region, allows Bub1 to interact with its partner Bub3 and to bind, together with Bub3, to MELT repeats, phosphorylated sequences located on Knl1. I found that this interaction controls the recruitment process to kinetochores and is critical for SAC activity. Additionally, the TPR of Bub1 establishes a second interaction with Knl1 to enhance Bub1 association with MELT repeats. My work describes the mechanism controlling Bub1 recruitment to kinetochores and paves the way for a molecular understanding of SAC signalling.

Federica La Mastra


Polycomb Repressive Complex 1 controls peripheral B cell homeostasis and terminal differentiation
Polycomb Repressive Complex 1 and 2 (PRC1 and PRC2) modulate chromatin accessibility through covalent histone modifications. The ubiquitous expression of PRC1 catalytic subunits throughout B cell development hypothesized roles for PRC1 in B cell physiology. This study aimed to dissect the contribution of PRC1 to peripheral B cell maturation, homeostasis and terminal differentiation. We analyzed mutant mice allowing conditional Cre-dependent inactivation of PRC1 catalytic function starting from late transitional B cells. In response to induced PRC1 inactivation, peripheral B cells in secondary lymphoid organs were reduced in number and displayed alterations in the surface phenotype, which reflected a major disturbance of their transcriptional profile. Reduced fitness of PRC1 mutant resting B cells was associated to heightened sensitivity to pro-apoptotic signals, consequent of the higher levels in these cells of the BIM protein and of the sub-optimal activation of the AKT kinase in response to either BAFF-R or BCR engagement. PRC1 mutant mice displayed major defects in the marginal zone (MZ) B cell subset, both in number and localization. This phenotype correlated with reduced expression of the Sphingosine-1-phosphate receptor-1 (S1pr1), which is crucial for B cell migration to the MZ, and the increased expression of the Polycomb target microRNA mir-125b, which targets S1pr1 transcript. Moreover, PRC1 mutant B cells displayed premature de-repression of Prdm1 gene and facilitated plasma cell differentiation upon lipopolysaccharide stimulation. Our work provides evidence for a crucial role played by PRC1 in peripheral B cell subset differentiation, B cell homeostasis and timing of terminal B cell differentiation.

Alma Linkeviciute


Cancer during pregnancy: a framework for ethical care
Cancer affects about 1 in 1000 pregnancies, which is estimated to account for approximately 5000 pregnancies in Europe each year. Despite affluent availability of clinical practice guidelines for cancer management in the course of pregnancy, there is very little ethical guidance consolidated for everyday use. Therefore, the aim of this work is to construct a framework, which would serve healthcare professionals as a reference tool when addressing ethical issues in cancer care during pregnancy. The framework is constructed around relational approach to respect for patient’s autonomy and ethics of care. It combines classical biomedical ethics principles and relational approaches to patient care. It emphasizes the importance of recognising the patient as another human being with her views on life, relationships and wishes, as well as forming a relationship between the patient and healthcare team, which is known as relational ethics. However, respect for individual autonomy is not discarded entirely. First and foremost, evidence-based information disclosure is essential to personalised patient care. The latter invites to integrate technical aspects of personalised medicine with relational approach to patient care. Furthermore, protection of the vulnerable also serves as a safeguard ensuring that patient is not compelled to choices that are not her own. Overall, this framework is expected to serve as a tool supporting ethical decisionmaking in cancer care during pregnancy. It can also be utilised by a variety of patient counselling services.

Pietro Lo Riso


Induced Pluripotent Stem Cells: An Innovative Tool to Dissect Ovarian Cancer Pathogenesis
Ovarian cancer (OC) has one of the highest death-to-incidence ratios among all tumor types, which points to the need for novel therapeutic and prognostic strategies. Indeed, the absence of relevant tumor cell lines that can recapitulate disease histopathology highlights an acute need for new model systems to study this pathology. In particular, it is still unclear whether the most common and aggressive form of this disease, high grade serous ovarian cancer (HGSOC), could arise from in the ovarian surface epithelium (OSE), as initially thought, or might be arising from the fimbrial epithelium. Here I addressed these issues in two complementary ways based on induced pluripotent stem cells: i) the modeling of Ovarian Cancer by somatic cell reprogramming to pluripotency of tumor cells; ii) the molecular characterization of HGSOC and its putative cells of origin. Somatic cell reprogramming, by erasing tumor-associated epigenetic marks while preserving the underlying genetic mutations, would allow for the first time the precise dissection of genetic and epigenetic contribution to this disease, through the differentiation of OC-iPSC into disease-relevant cell types. I demonstrated the feasibility of OC reprogramming through a non-integrative platform, showing that OC-derived iPSC are closely similar to human ESC, and proving their tumoral origin by whole exome sequencing. Moreover, I showed that independent iPSC clones derived from the same tumor upon trilineage differentiation in vivo show differential tumorigenic potential. For a more precise dissection of this phenotype, I set up a differentiation protocol 13 that allows differentiation of pluripotent cells into mesodermal progenitors, that are precursors of both fimbria and OSE. To isolate a pure population of these cells, I resorted to CRISPR/Cas9 to integrate a selection cassette in the MIXL1 locus. By this approach, I was able to show correct gene targeting at the intended site, allowing also for selection of mesodermal progenitors upon differentiation of normal iPSC. The same approach translated to OC-derived iPSC would allow to study the effects of genetic mutations deprived of tumorassociated epigenetic marks during differentiation, both at the stage of mesodermal progenitors and in cells directed towards the female reproductive epithelium in vivo. The second approach relies on the identification of specific molecular features of fimbria and ovarian surface epithelium, the two putative cells of origin of HGSOC. On this side, I offer a first glimpse on molecular features of HGSOC cancer and normal gynecological tissues. I could show that specific DNA methylation signatures of fimbrial epithelial cells and ovarian surface epithelium cells are partially retained in tumor samples and stratify HGSOC samples according to the putative cell of origin of this tumor. Moreover, I show for the first time a description of histone modifications in primary HGSOC, concentrating on marks of activation/repression sitting on promoter regions (H3K4me3 and H3K27me3, respectively) and marks that characterize active/closed-poised enhancers (H3K4me1, H3K27ac and H3K27me3).

Chiara Locarno


Proteomic identification of the transcription factors Ikaros and Aiolos as new Myc interactors on chromatin
Myc is a transcription factor that plays a key role in many cellular functions. When deregulated, it becomes a potent oncogene and a hallmark of many human cancers. Although its activity has been proven to strongly depend on many cofactors, we are still far from a comprehensive understanding of Myc’s mechanisms of action in transcriptional control. Moreover, since Myc itself is not easily druggable, targeting its cofactors or downstream effectors may constitute a more promising therapeutic strategy in Myc-driven tumors. Here, we have used the Chromatin Proteomics (ChroP) technology to identify new Myc interactors. Among the list of identified candidates, we have chosen to focus our attention on two lymphocyte specific transcription factors, Ikaros and Aiolos. We validated the interaction of Myc with these factors and started to study its functional meaning. Our working hypothesis is that Myc and Ikaros/Aiolos antagonize each other on chromatin at selected target loci. We were able to identify a subset of genes regulated in opposite manners by Ikaros and Myc, involved mainly in cell cycle regulation, metabolism of lipids and metabolic stress response; the next step will be to study more in depth the mechanisms behind their regulation and their functional significance.

Sara Loponte


Histone deacetylases (HDACs) are a class of modification enzymes that catalyze the removal of acetyl molecules from histone and non-­‐histone substrates. Therefore, they play important roles in chromatin remodelling and gene expression control through
Histone deacetylases (HDACs) are a class of modification enzymes that catalyze the removal of acetyl molecules from histone and non-­‐histone substrates. Therefore, they play important roles in chromatin remodelling and gene expression control through regulation of histones, transcription factors, and chromatin-­‐modifying enzymes. Class I HDACs, in particular HDAC1 and HDAC2, are ubiquitously expressed and are,critical regulators of cell cycle progression, cellular proliferation and differentiation during development. Besides their subcellular localization and incorporation into multi-­‐0subunit complexes, HDAC1 and HDAC2 can also be regulated by a plethora of post-­‐translational modifications (PTM), which represent a complex “code” that modulates their catalytic activity, localization and complex assembly. Among the various PTMs that occur on HDAC1 and HDAC2 we identified a new mitotic-­‐specific phosphorylation of the two enzymes driven by Aurora kinases A and B. By means of mammalian cells and zebrafish embryos, we dissected the biological role of HDAC1 Aurora-­‐dependent phosphorylation during development, contributing to deciphering the PTM code of these deacetylases. Indeed, we demonstrated that this phosphorylation in vitro and in vivo partially and dynamically affects HDAC1 enzymatic activity. Moreover, we found that Aurora-­driven phosphorylation of HDAC1 is critical for the maintenance of a proper proliferative and developmental plan in a complex organism and crucially regulates global histone acetylation levels in zebrafish development. Thus, affecting its activity on histone acetylation, HDAC1 mitotic phosphorylation acts as fine-­tune regulator of proper cell cycle progression, probably by modulating the expression of genes directly involved in zebrafish development.

Caterina Lucano


A novel role of the endocytic adaptor proteins Eps15 and Eps15L1 in the regulation of Notch signaling
Notch signaling is an evolutionary conserved signaling pathway that regulates multiple aspects of development and cell renewal. Endocytosis plays a critical role in Notch signaling regulation, both in the signal-sending and in the signal-receiving cell. Eps15 and Eps15L1 are two endocytic adaptors, involved in clathrin-dependent and independent endocytosis of tyrosine kinase receptors. Double Knockout (DKO) mice for Eps15/15L1 die between 9.5 and 11.5 dpc and show a Notch loss of function phenotype, mirrored by downregulation of Notch target genes. Based on these observations we decided to test whether Eps15 and Eps15L1 can regulate Notch signaling. Using an in vitro coculture/transactivation assay we observed no reduction in Notch activity after knockdown (KD) of Eps15 or Eps15L1 in the signal-receiving cell. However, after KD of Eps15 or Eps15L1 in the signal-sending cell a 40-50% reduction in Notch activity was observed, even though no further decrease of the signal was observed after the combined Eps15/L1 KD. This observation was true for the four Notch ligands Dll1, Dll4, Jag1 and Jag2. To understand the molecular mechanism underlying this regulation, we set up a FACS based Dll1 internalization assay. While we observed a strong reduction in Dll1 endocytic constant following KD of known endocytic regulators of Notch ligands, no reduction was observed after KD of Eps15, Eps15L1 or both. Moreover, we assessed that localization of Dll1 in detergent resistant membrane was not altered following Eps15/L1 KD. At the moment we are setting up a quantitative Dll1 recycling assay and a structure/function rescue with Eps15 mutants to understand other possible mechanisms through which Eps15/L1 can regulate Notch ligands activity.

Eleonora Lusito


A Network-based Approach to Breast Cancer Systems Medicine
Breast cancer is the most commonly diagnosed cancer and the second leading cause of cancer death in women. Although recent improvements in the preven- tion, early detection, and treatment of breast cancer have led to a significant decrease in the mortality rate, the identification of an optimal therapeutic strat- egy for each patient remains a difficult task because of the heterogeneous nature of the disease. Clinical heterogeneity of breast cancer is in part explained by the vast genetic and molecular heterogeneity of this disease, which is now emerging from large-scale screening studies using “-omics” technologies (e.g. microarray gene expression profiling, next-generation sequencing). This genetic and molecu- lar heterogeneity likely contributes significantly to therapy response and clinical outcome. The recent advances in our understanding of the molecular nature of breast cancer due, in particular, to the explosion of high-throughput technolo- gies, is driving a shift away from the “one-dose-fits-all” paradigm in healthcare, to the new “Personalized Cancer Care” paradigm. The aim of “Personalized Cancer Care” is to select the optimal course of clinical intervention for individ- ual patients, maximizing the likelihood of effective treatment and reducing the probability of adverse drug reactions, according to the molecular features of the patient. In light to this medical scenario, the aim of this project is to identify novel molecular mechanisms that are altered in breast cancer through the devel- opment of a computational pipeline, in order to propose putative biomarkers and druggable target genes for the personalized management of patients. Through the application of a Systems Biology approach to reverse engineer Gene Regula- tory Networks (GRNs) from gene expression data, we built GRNs around “hub” genes transcriptionally correlating with clinical-pathological features associated with breast tumor expression profiles. The relevance of the GRNs as putative cancer-related mechanisms was reinforced by the occurrence of mutational events related to breast cancer in the “hub” genes, as well as in the neighbor genes. Moreover, for some networks, we observed mutually exclusive mutational pat- terns in the neighbors genes, thus supporting their predicted role as oncogenic mechanisms. Strikingly, a substantial fraction of GRNs were overexpressed in triple negative breast cancer patients who acquired resistance to therapy, suggesting the involvement of these networks in mechanisms of chemoresistance. In conclusion, our approach allowed us to identify cancer molecular mechanisms frequently altered in breast cancer and in chemorefractory tumors, which may suggest novel cancer biomarkers and potential drug targets for the development of more effective therapeutic strategies in metastatic breast cancer patients.

Elisa Magistrati


Functional characterization of myosin VI in centrosome biology and cell cycle progression
Myosin VI is a unique actin motor involved in multiple biological functions, such as endocytic and secretion processes, cell migration, autophagy, and in the maintenance of the Golgi complex and stereocilia. These functions are dictated by the interaction of myosin VI with different cargos, which can also regulate the ability of this protein to work as an anchor or a motor that moves along actin filaments. Previous experiments performed in our laboratory led to the identification of a novel set of myosin VI interactors that belong to the centrosome compartment, suggesting that myosin VI could have an important and unexpected role in centrosomal processes. Indeed, the depletion of myosin VI leads to alterations in the centrosome structure and number, and to an impairment in the formation of the primary cilium. Our data also suggest a potential role of myosin VI in the regulation of cell cycle progression. Indeed, myosin VI depletion leads to cell cycle arrest and senescence caused by p53 activation. To characterize this phenotype, we performed a genome-wide CRISPR/Cas9 rescue screening, that led us to identify some candidates potentially involved in p53 activation following myosin VI depletion. This study unveils a new role for myosin VI in centrosome biology and in the control of the cell cycle, two processes whose dysregulation is an important step during carcinogenesis.

Chiara Malinverno


RAB5A in the control of mammary epithelial morphogenesis and motility
RAB5A, a master regulator of endocytosis, promotes a tumor mesenchymal invasive program. In Drosophila, however, loss-­‐of-­‐function mutant of RAB5 transforms imaginal disc epithelia into highly proliferative tissues, pointing to a tumor suppressor function. To dissect the complex role of RAB5A in tumor development, we investigated its impact on MCF-­10A, an immortalized non-­‐transformed mammary epithelial cell line that mimics morphogenesis of mammary gland when culture on 3D reconstituted basement membrane. We generated inducible MCF-­‐10A cells expressing either RAB5A-­WT or its dominant negative form (RAB5A-­S34N). We found that the expression of RAB5A-­S34N is sufficient to sustain MCF-­10A cells proliferation in the absence of EGF, through the secretion of a diffusible growth-­‐ promoting factor. Conversely, the expression of RAB5A-­WT delayed cell cycle progression of cells grown in 2D, albeit it promoted the formation of hyperproliferative acini when grown in 3D. Thus, RAB5A may either be implicated in growth factor independent growth, or may promote proliferation in 3D. Both clinical data and in vitro studies demonstrated that RAB5A is required for invasion and metastasis, suggesting its involvement in tumor progression. To further explore this latter role, we tested MCF-­10A cells motility. We demonstrated that RAB5A expression does not affect single cell migration, but specifically enhances collective locomotion. Indeed, RAB5A-­expression promotes increased coordination and coherence of epithelial cell sheet motility, related both to the increase in the area and persistence of cell protrusions at the leading front, and possibly to a tightening of cell-­‐cell contacts. We are currently dissecting the molecular mechanisms through which RAB5A altered function or expression impact on both mammary gland morphogenesis and tumor progression.

Maria Mallardo


Dissecting the role of the cytoplasmic mutant Nucleophosmin in acute myeloid leukaemia development
Acute myeloid leukaemia (AML) is a genetic heterogeneous group of diseases, with the largest subgroup showing a mutation in the Nucleophosmin gene (NPM1). Normally the NPM protein localizes mainly in the nucleolus, but in AML blasts it is aberrant localized to the cytoplasm (NPMc+AML). Notably, NPMc+AML patients show peculiar gene expression profiles, treatment response and prognosis. Hence, it has been proposed as an independent category for leukaemia classification according to WHO in 2008. In view of the relevance of NPMc+ mutation to AML pathogenesis and prognosis, understanding its role in leukaemia development represents a major issue in the field. The aim of this PhD project is to get further insight into the relevance of NPMc+ mutations to AML development. To this scope, here it is reported a characterization of a novel mouse model expressing the mutated protein. The hematopoietic restricted expression of the protein induces leukaemia in mice. This data definitively clarify that NPMc+ is an initiating mutation for leukaemia development. However, the long latency and low penetrance of disease onset strongly support the need of cooperating mutations. Since, the high frequency of FLT3-ITD mutations in NPMc+AML, we genetically tested the synergisms between the two abnormalities. To this scope, NPMc+ mice were crossed with FLT3-ITD mice (Lee, 2007). Double mutated mice developed leukaemia with sort latency and full penetrance indicating effective cooperation. Moreover, our data support the two hits model of tumourigenesis, where functional complementary mutations contribute to disease onset. Another major challenge of this project is to understand how NPMc+ affect the biology of normal HSPC and imposes the transition from normal to cancer stem cells. We found that NPMc+ expression perturbs the homeostasis of HSCP and expand the number of LT-HSC by increasing the proliferation rate. However, this enhanced proliferation is not associated to loss of quiescent and functional HSC, which may represent a reservoir of persistent pre-malignant cells available for the accumulation of additional genetic alteration. Further investigation into the biology of per-leukaemic stem cells may give insights into the molecular mechanisms imposed by the oncogene for malignancy transformation and finally may contribute for the development of new therapeutic strategies.

Luca Marelli


The co-production of scientific and Translational induced pluripotent stem Reprogramming platforms. Governance innovation in Cell research
This dissertation charts the rise and articulation of induced Pluripotent Stem Cells (iPSCs) as a prominent translational technology, invested with high expectations to finally deliver the as yet mostly unfulfilled promise of stem cell research. In a field catalyzed by the therapeutic promise, iPSCs have been adopted for widespread translational efforts, in the areas of disease modeling, drug discovery and regenerative medicine, and have progressively positioned themselves, through the mobilization of several biomedical platforms, as a key resource of stem cell-based bioeconomies. Specifically, drawing from extensive ethnographic fieldwork, this work targets distinct iPSC innovation pathways across the United States and the European Union, and conducts the analysis of distinct models of iPSC–based innovation implemented by three leading iPSC research organizations that have been spearheading translational iPSC research: the New York Stem Cell Foundation, the Harvard Stem Cell Institute, and the European Bank for induced Pluripotent Stem Cells – respectively, the largest stem cell research organization in the world; the largest private translational stem cell research institution in the United States; and one of the two flagship stem cell consortia launched in recent years at EU level. Through a comparative approach, this dissertation explores the co-productive relationship between scientific and governance innovation, and probes the distinct ways in which some of the leading research institutions in the field design and implement governance arrangements and practices of standardization in order to harness the innovation potential of iPSC-based technologies. Furthermore, it accounts for the socio-political salience of these emerging institutional configurations, and traces the assembly of distinct constituencies claiming jurisdiction in this domain of biomedicine.

Lucia Massari


Complete resolution of sister chromatid intertwines requires the Polo-like kinase Cdc5 and the phosphatase Cdc14 in budding yeast
During mitosis the newly replicated genetic material, organized in sister chromatids, is equally subdivided into the daughter cells through a fine-regulated process called chromosome segregation. Sister chromatids are held together and identified as sisters by cohesin. At the metaphase-to-anaphase transition, when all chromatids are correctly attached to the spindle, cohesin is cleaved and chromosome segregation initiates. Beside cohesin, all linkages between sister chromatids need to be removed to allow for their complete separation. Additional linkages include DNA linkages (or sister chromatid intertwines, SCIs), such as recombination intermediates and DNA catenanes. In Saccharomyces cerevisiae a mutant that lacks the activities of the Polo-like kinase Cdc5 and the phosphatase Cdc14, two major mitotic regulators, has been identified that proved to be particularly suitable for studying SCIs that persist in mitosis. The cdc5 cdc14 double mutant arrests with short and stable mitotic spindles and unseparated nuclei, despite having cleaved cohesin. In addition to having a spindle elongation defect, these cells are also impaired in the resolution of cohesin-independent linkages between chromatids. We found that these linkages mostly consist of DNA catenanes, that persist in cdc5 cdc14 cells at their terminal arrest and that are sufficient to counteract spindle elongation. Our results suggest that Cdc5 is required for their resolution. This finding, together with the knowledge that Cdc5 promotes Cdc14 activation and that both proteins are essential for spindle elongation and mitotic exit, allows us to speculate that they coordinate different aspects of chromosome segregation to guarantee genome integrity throughout mitosis.

Valeria Mastrodonato


The study of Snap29 in mitosis and in CEDNIK pathogenesis
The fusion between a vesicle and a target membrane is a key step of intracellular trafficking mediated by SNARE proteins. Snap29 is a cytosolic SNARE protein, whose specificity and activity is unclear. During the last few years, Snap29 was discovered to be a key regulator of autophagy required for fusion of autophagosomes with lysosomes. During my PhD, I contributed to uncover a novel function of Snap29 in Drosophila, demonstrating that during mitosis Snap29 acts as a kinetochore component and it is required to ensure proper cell division. Indeed, depletion of Snap29 in S2 cells determines defective chromosome segregation, often leading to cell death. In addition, we observed that Snap29 ensures correct tissue development and homeostasis in Drosophila, since its depletion or mutation causes epithelial multilayering and tumor-like tissue alterations. Since mutations affecting autophagy genes are not sufficient per se to induce such disruptions, we hypothesize that these defects might be also due to loss of Snap29 activity during mitosis. Mutations of SNAP29 human gene cause a rare syndrome called CEDNIK, which leads to severe neurological and dermatological manifestations So far, the most investigated aspects of this syndrome are dermatological alterations, likely caused by the impairment of SNAP29 activity during membrane trafficking. To study uncharacterized CEDNIK traits, we took advantage of a zebrafish snap29 mutant and we found that, beside known CEDNIK symptoms, they display trigeminal nerve formation and axonal branching defects, suggesting the requirement of Snap29 for correct nervous system development. Overall, our findings demonstrate that Snap29 is a key regulator of cell division and shed light on uncharacterized aspects of CEDNIK syndrome, suggesting a pivotal role of Snap29 in nervous system development.

Parinaz Mehdipour


Dissecting the role of histone deacetylase 3 (HDAC3) in leukemogenesis
Acute Promyelocytic Leukemia (APL) is the first model disease in which the involvement of HDACs has been documented. In this study we characterized the role of HDAC3 through a functional knock-down approach, assessing its impact on cellular differentiation, proliferation and the ability to influence the transplantation of HSCs and APL cells. Indeed, Hdac3-KD in vitro reduced the proliferative potential of both pre-leukemic and full leukemic cells and boosted their differentiation, suggesting that HDAC3 plays the role of an oncogene in APL initiation and progression. These results were not restricted to APL, because lymphoma driven by c-myc overexpression and leukemia driven by MLL-AF9, were both impaired in cell growth upon Hdac3-KD. In vivo, inoculation of Hdac3-KD pre-leukemic cells into lethally irradiated recipient mice or inoculation of Hdac3-KD APL cells into the recipient mice did not result in leukemia development or progression, respectively. These results suggest that HDAC3 can be considered as a target for epidrugs in the treatment of hematological malignancies. Thus, we assessed this hypothesis with the treatment of pre-leukemic and leukemic cells with the HDAC3 selective inhibitor, RGFP966. Indeed, inhibition of HDAC3 enzymatic activity with RGFP966, phenocopied Hdac3- KD phenotypes in pre-leukemic and leukemic cells confirming the putative oncogenic role of HDAC3. In conclusion, my PhD project has expanded our comprehension about the role of HDAC3 in hematological malignancies and is beginning to unravel alternative views on the targets of epidrugs for the treatment of leukemic patients.

Alessia Melacarne


Identification of neoantigens released by Salmonella-infected tumor cells for a novel approach to cancer immunotherapy
Advancement in cancer immunotherapy has revealed the importance of targeting neoantigens: tumor specific antigens that prompt a strong antitumor response escaping from the central T tolerance. Our laboratory has previously shown that infection of mouse tumor cell lines with Salmonella induces the transfer of antigens between adjacent cells, specifically through hemichannels. Now we demonstrate that Salmonella not only leads to the transfer of immunogenic antigens between adjacent cells but also the release of peptides in the extracellular milieu. A mass spectrometry analysis of proteins and peptides released by infected murine melanoma cells revealed that among them there are novel tumor epitopes and potential neoantigens. We tested the released peptides as components of a vaccine formulation given to prevent tumor progression in a murine model of melanoma and we demonstrated that are immunogenic. We attested that also human melanoma cell lines infected with Salmonella release peptides; these induce the expansion of CD8-T cell from peripheral blood mononuclear cells that specifically kill human melanoma cells in vitro. We propose the treatment with Salmonella of patients-derived tumor cells as a strategy to obtain neoantigens that could be applied in clinical studies as a vaccine formulation. Encouraging preliminary results have been obtained with the therapeutic treatment of pet dogs with spontaneous osteosarcoma and high grade sarcoma with a vaccine formulation based on peptides released by their own tumor cells following Salmonella infection.  

Giorgio Melloni


Computational frameworks for the identification of somatic and germline variants contributing to cancer predisposition and development
The most recent cancer classification from NIH includes ~200 types of tumor that originates from several tissue types (http://www.cancer.gov/types). Although macroscopic and microscopic characteristics varies significantly across subtypes, the starting point of every cancer is believed to be a single cell that acquires DNA somatic alterations that increases its fitness over the surrounding cells and makes it behave abnormally and proliferate uncontrollably. Somatic mutations are the consequence of many possible defective processes such as replication deficiencies, exposure to carcinogens, or DNA repair machinery faults. Mutation development is a random and mostly natural process that frequently happens in every cell of an individual. Only the acquisition of a series of subtype-specific alterations, including also larger aberrations such as translocations or deletions, can lead to the development of the disease and this is a long process for the majority of adult tumor types. However, genetic predisposition for certain cancer types is epidemiologically well established. In fact, several cancer predisposing genes where identified in the last 30 years with various technologies but they characterize only a small fraction of familial cases. This work will therefore cover two main steps of cancer genetics and genomics: the identification of the genes that somatically changes the behavior of a normal human cell to a cancer cell and the genetic variants that increase risk of cancer development. The use of publicly available datasets is common to all the three results sections that compose this work. In particular, we took advantage of several whole exome sequencing databases (WES) for the identification of both driver mutations and driver variants. In particular, the use of WES in cancer predisposition analysis represents one of the few attempts of performing such analysis on genome-wide sequencing germline data.

Valentina Melocchi


A novel approach for the identification of candidate driver lesions in breast cancer based on the comparison of the mutational profiles of a primary tumour and its matched mammospheres and xenograft
The clinical management of breast cancer patients is complicated by the high genetic heterogeneity of this disease, which makes the standardization of treatments, the prediction of prognosis and therapy response, and the development of personalized therapies difficult. Nevertheless, the advent of high-throughput genomics screenings based on microarray or next-generation sequencing (NGS) technologies has greatly enhanced our understanding of the genomic landscapes underlying breast cancer development and progression. Such discoveries are now allowing clinicians to tailor therapies based on the molecular subtype of the tumour (luminal, basal and HER2). NGS studies have also started to provide insights into the range of molecular profiles of tumour cells from the same tumour, and have shown that in some breast cancers a high level of intratumoral genetic heterogeneity exists. The findings from these studies support a scenario in which breast tumours can be either: i) monogenomic, comprised of a single clonal cell population; ii) or polygenomic, composed of several related clonal subpopulations. The co-existence of different cancer driver genetic lesions in polygenomic tumours might contribute to treatment failure in some cases, as relapse could be driven by the expansion of a subpopulation of cells intrinsically resistant to the therapy. Importantly, cancer genetic heterogeneity has been recapitulated in experimental settings using cancer stem cells (CSCs) xenografted in mouse models. We hypothesized that the mutational events that drive the onset and progression of breast tumours lie within the CSC compartment. To explore this possibility, we analysed and compared the mutational profiles of a primary breast tumour and its matched mammospheres (source of CSC-derived population), patient-derived xenograft (PDX) and PDX-derived mammospheres using Whole Exome Sequencing (WES). We setup a NGS approach to look for rare mutations in the primary tumour that may be present in the CSC compartment using low amounts of DNA input. We optimised an experimental protocol in which the genomic DNA (gDNA) of each sample was subjected to Whole Genome Amplification (WGA) prior to performing WES. This enabled us to obtain a sufficient amount of DNA (≥ 3 μg) to perform WES. We also introduced a filtering step in our analysis, based on the Xenome software, for PDX-derived samples to eliminate possible contamination from murine DNA. Our study allowed us to characterize the genetic profiles of CSCs and to identify cancer-relevant mutations that could drive breast cancer onset and progression. We identified 15 candidate driver mutations in 11 genes that were enriched, in terms of mutation frequency, within primary tumour-derived mammospheres and the PDX. Together with these mutations, we identified 4 mutations in 4 genes, not enriched, but shared among all analysed samples, which likely represent “founder” mutations. Based on our results, we will now endeavour to determine the clinical relevance of the candidate driver mutations identified in our study by determining their prevalence in independent patient cohorts. Having optimised the protocol for NGS of matched primary tumour, PDX and mammosphere populations, we will also extend our mutational analysis to additional breast tumours for the identification of more driver mutations and for the deconvolution of intratumoral genetic heterogeneity of breast cancer. Understanding the driving mutational forces of breast tumours and relative mechanisms involved is paramount for the development of more effective therapeutic strategies.

Demis Menolfi


Essential postreplicative functions of the Smc5/6 complex
The structural maintenance of chromosomes (SMC) complex Smc5/6 is based on a heterodimer of two SMC subunits, Smc5 and Smc6, and six non-Smc element subunits, Nse1-6, all of which are essential for cell viability in most organisms. Smc5/6 safeguards genome integrity via different mechanisms, including stabilization of stalled replication forks, resolution of recombination intermediates, and maintenance of nucleolar integrity. However, the essential functions of Smc5/6 remain elusive. The aim of the present work was to understand when in the cell cycle the crucial functions of Smc5/6 are manifested and to identify them. Through the use of cell cycle regulated alleles, which enabled the restriction of various Smc5/6 subunits expression to either S or G2/M phases of the cell cycle, we uncovered that the essential roles are executed postreplicatively in G2/M. By further genetic screens, molecular approaches and genome-wide studies, we identified three chromosome topology and recombination-related processes that are crucially sensitive to low amounts of Smc5/6 specifically in G2/M. First, Smc5/6 plays a topological role affecting the formation and/or the resolution of Rad5-Mms2-Ubc13 chromatin structures that are later engaged by Sgs1-Top3-Rmi1. Second, Smc5/6 facilitates an epigenetic pathway that ensures silencing of specific loci, such as repetitive DNA regions, thereby preventing unrestrained recombination. Third, Smc5/6 has an anti-fragility function, facilitating replication through natural pausing elements and site-specific replication fork barriers and preventing their breakage in mitosis during chromosome segregation.

Flavia Michelini


A new class of non-coding RNA Controls the DNA damage Response and DNA repair
The DNA damage response (DDR) is a signaling cascade that follows the generation of a lesion inthe DNA double helix and promptly arrests cell proliferation in order to attempt DNA repair. Noncoding RNAs have been involved in an increasing number of cellular events and some of them are processed by members of the RNA interference (RNAi) pathway. During my PhD, I contributed to uncover an unexpected layer of DDR regulation by a new class of DICER- and DROSHAdependent small non-coding RNA, named DDRNA. I demonstrated that DDR foci stability is sensitive to RNA polymerase II inhibition and to RNase A treatment. Incubation of RNase A-treated cells with DICER- and DROSHA-dependent RNA products restores focal accumulation of DDR factors. DICER and DROSHA are indeed necessary to trigger DDR upon exogenous DNA damage in human cells, in a miRNA-independent manner. In a mammalian cell system in which a single DNA double-strand break can be generated at a defined locus, DDR focus formation requires site-specific RNA molecules. RNA deep sequencing confirmed the presence of DICER-dependent 22-23-nucleotide transcripts arising from the damaged locus. These DDR-regulating RNAs (DDRNAs) act at the first steps of the DDR cascade, in an MRN-dependent manner. Importantly, DDRNAs, both chemically synthesized or generated in vitro by DICER cleavage, are biologically active. Finally, fluorescently labeled DDRNAs have been demonstrated to localize site-specifically at the damaged locus. Collectively these results suggest an unanticipated direct role of DICER and DROSHA in the production of small non-coding RNAs that control DDR activation at sites of DNA damage.

Marta Milan


Involvement of the transcription factor MYRF in signaling from the endoplasmic reticulum to the nucleus in pancreatic cancer
Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of cancer death worldwide. One reason for the poor prognosis is the high intra-tumor heterogeneity, with the coexistence of well- and poorly-differentiated cells in virtually all tumor cases. Thus, a better characterization of the circuitries regulating PDAC cell differentiation is required. We found that MYRF, a poorly characterized transcription factor, is selectively expressed in well-differentiated PDAC cell lines. MYRF is synthetized as an endoplasmic reticulum (ER) membrane protein and self-cleaves after trimerization, releasing the N-terminal trimer that translocates into the nucleus and regulates transcription. We generated MYRF-KO PDAC cells and combined transcriptomic profiles and analyses of MYRF genomic occupancy to study its function. We retrieved the MYRF DNA binding motif from our ChIP- sequencing data and demonstrated that MYRF capability to bind this sequence and activate transcription is strictly dependent on its trimerization. MYRF deletion resulted in the downregulation of cell replication-related genes and upregulation of ER stress-related genes. Consistently, MYRF loss resulted in an altered ER morphology and function, probably as a result of the overexpression of membrane and secreted proteins with complex folding, such as cysteine rich and highly glycosylated proteins. Additionally, we found that MYRF creates a feed-forward loop with the transcription factors FOS and FOSB. In doing so, MYRF directly regulates the expression of these two transcription factors that in turn cooperate to generate the MYRF transcriptional outcome. In conclusion, this work points to a role for MYRF as a key player in the communication between ER and nucleus, working as a sensor of proper ER function. MYRF appears to license cells for DNA replication and concomitantly to serve as a guardian against ER overload in highly secretory cancer cells.

Cinzia Milesi


Redundant And Non-Redundant Roles Of The Endocytic Adaptor Proteins Eps15 And Eps15l1 In Mammals
Eps15 and Eps15L1 are two homologous endocytic adaptor proteins that have been involved in endocytosis of several plasma membrane receptors through in vitro studies on cell lines. Eps15 knockout (KO) mice are healthy and fertile, while Eps15L1-KO mice die at birth presumably because of neural defects. Importantly, Eps15/Eps15L1 double knockout (DKO) mice have a more severe phenotype, dying at midgestation, suggesting redundancy of the two proteins in one or more fundamental developmental programs. The aim of this thesis project was to investigate redundant and non-redundant roles of Eps15 and Eps15L1, with the final goal to unmask the underlying causes of embryonic lethality of Eps15/Eps15L1-DKO mice. Since Eps15/Eps15L1-DKO mice displayed a Notch loss-of-function phenotype, we hypothesized that Eps15 and Eps15L1 redundantly regulated Notch signalling. Using mouse embryonic fibroblasts as signal-sending cells in a co-culture model system, we found that Eps15L1 non-redundantly activated Notch signalling. This finding indicated that impaired Notch signalling only partially contributed to the DKO phenotype. Since Eps15/Eps15L1-DKO mice were pale, we hypothesized that Eps15 and Eps15L1 redundantly regulated erythropoiesis. Analysis of conditional Eps15/Eps15L1-DKO mice, lacking Eps15 and Eps15L1 in hematopoietic cells, revealed that these mice survived to adulthood but suffered from microcytic hypochromic anemia. Our data suggest that impaired erythropoiesis might be linked to a redundant role of Eps15 and Eps15L1 in the regulation of endocytosis of the transferrin receptor in red blood cells. Our findings indicate that Eps15 and Eps15L1 regulate several developmental programs, some non-redundantly and others redundantly, possibly addictively contributing to the DKO phenotype.

Simona Monterisi


HOXB7 in lung cancer: a novel role in stem cell and iPS biology
Current diagnostic tools do not allow prognostic evaluation of patients with early stage lung cancer or selection of patients that might bene t from adjuvant chemotherapy. Therefore, the identi cation of novel prognostic markers in early-stage lung cancer is paramount. In this scenario, the transcription factor HOXB7, belonging to the homeobox family, has been shown to correlate with poor prognosis in di erent types of cancer and recently also in stage I lung adenocarcinoma. To better understand the prognostic implication of alterations in HOXB7 expression in lung cancer, we performed a bioinformatics analysis of multiple lung cancer expression datasets in order to identify gene sets representing cancer-relevant biological functions enriched in high-HOXB7 expressing tumors. We found several gene sets enriched in high-HOXB7 expressing tumors representing molecular mechanisms involved in epithelial to mesenchymal transition, in cancer progression, and, interestingly, in stemness and cellular reprogramming. Based on these results, we hypothesized that HOXB7 may have a role in the expansion of the stem cell compartment in cancer, a mechanism that has been shown to be a hallmark of enhanced tumorigenicity and of increased metastatic potential. Analysis of the stem-related surface marker CD90 revealed that overexpression of HOXB7 in lung cells increases the CD90high sub population. CD90high, but not CD90low cells, are able to form spheroids, which is an hallmark of stemness. Indeed, the sphere forming eciency of normal lung BEAS-2B cells was 22% and 1.64% in CD90high and CD90low populations, respectively. In addition, we found that silencing of LIN28B counteracts the expansion of the CD90high population. LIN28B was recently described as an oncogene that regulates the cancer stem cell compartment. We found that LIN28B is under the direct transcriptional control of HOXB7. Therefore, we propose a novel molecular mechanism driven by HOXB7 and can increase stem-like properties in lung cells. We further demonstrated that the HOXB7-LIN28B axis plays an important role in reprogramming of adult cells into induced pluripotent stem cells (iPS). Indeed, HOXB7 may enhance the reprogramming eciency achieved by the three genes OCT4, KLF4, SOX2 in both mouse embryonic broblast and human epithelial BEAS-2B cells by substituting MYC in the transcription factor cocktail of reprogramming factors used by Yamanaka. Of note, LIN28B silencing strongly decreases the number of reprogrammed colonies in high-HOXB7 expressing cells. These ndings suggest that HOXB7, through transcriptional induction of the LIN28B gene, activates a program relevant for stem/iPS cell biology and for tumor progression, possibly opening a new line of research for the development of more e ective therapies for metastatic lung cancer patients.

Elena Morelli


Novel functions of the SNARE protein Snap29 in membrane trafficking and cell division
Vesicular trafficking within cells is an important process for tissue development and homeostasis. We have isolated a loss of function mutant in the Drosophila homolog of snap29 gene, which encodes a ubiquitous SNARE protein (Snap29 hereafter) shown to be involved in membrane fusion. Lack of Snap29 functionality in Drosophila epithelial tissues correlates with defects of tissues architecture and development. Our findings support a role of Snap29 at key steps of membrane trafficking. In particular, we found that Snap29 is required for fusion of autophagosomes with lysosomes and that lack of Snap29 results in excess of secretion, suggesting that Snap29 might act negatively in regulation of vesicle fusion at the plasma membrane. Surprisingly, we found that both in Drosophila and mammalian cells Snap29 controls the attachment of mitotic fibers to kinetochores during mitosis. In mammalian cells, we show that lack of SNAP29 correlates with absence at kinetochores of ZWINT-1 and ZWILCH, a component of RZZ complex, with prometaphase delay and formation of daughter cells containing mininuclei. In vivo, we demonstrated that while autophagy defects are not the cause of the altered epithelial tissues architecture in snap29 mutants, defective mitotic phases are frequently found in snap29 mutant tissues suggesting that alterations of mitosis might be responsible for altered tissues phenotype. Our study contribute to shed light on the pathogenesis of CEDNIK, a human congenital syndrome caused by SNAP29 inactivation and we surmise that the function of SNAP29 at KT could be potentially relevant to development of aneuploidy in tumor-like masses.

Marco Morini


VE-cadherin orchestrates epigenetic modifications aimed at endothelial stabilisation
The mechanistic foundation of vascular stability is still largely unknown. A gene profile analysis reveals that the expression of the endothelial specific, adherens junction protein VE-cadherin upregulates a set of genes important in vascular stability that includes claudin-5, VE-PTP and von Willebrand factor. Mechanistic studies indicate that VE-cadherin exerts this effect by inhibiting the nuclear translocation of FoxO1 and b-catenin that, in turn, anchor the Polycomb multiprotein repressive complex (PRC2) to specific promoters. VE-cadherin also sequesters a core subunit of PRC2 (Ezh2) at the cell membrane thus preventing its nuclear translocation. Dismantling of VE-cadherin clusters during pathological angiogenesis in human ovarian carcinomas is accompanied by increase in EZH2 and down-regulation of claudin-5. These data introduce the novel concept of an inhibitory role of Polycomb proteins in EC differentiation and vascular stability and open novel therapeutic opportunities to induce vascular normalization through inhibition of the Polycomb-mediated repression system.

Joanna Niska


Terminating Replication at TERs at Eukaryotic Chromosomes
Faithful transmission of genetic material is challenged by the presence of natural impediments affecting replication fork progression that jeopardize genome integrity. Transcription, which competes with DNA replication for the same template, is a common barrier to replication in both prokaryotes and higher eukaryotes. Multiple mechanisms minimize the consequences of DNA replication and transcription collisions in order to prevent torsional stress accumulation that occurs when replication fork encounters the transcription machinery. Defects in resolving topological problems during chromosome replication lead to fork reversal, R loop formation and recombination-induced genome rearrangements. Our interest is focused on the processes that coordinate replication with transcription at TERs (termination sites) and on the molecular pathways involved in termination of DNA replication. We investigated the roles of Rrm3, a DNA helicase that assists replication fork progression, and of Sen1, a DNA/RNA helicase that resolves the conflicts between replication and transcription. We found that Rrm3 and Sen1 mediate replication termination at specific TERs, preventing aberrant events ultimately leading to chromosome fragility. Our results contribute to the elucidation of mechanisms coordinating replication and transcription at TER zones in eukaryotes.

Koustav Pal


Exploring changes in higher-order genome organisation during the coordinated transcriptional up- regulation in drosophila dosage compensation
Proper chromosome segregation requires an orderly sequence of events, whereby spindle elongation follows the dissolution of sister chromatid linkages. The mitotic spindle is a sophisticated and complex machinery built of microtubules, microtubule associated proteins and motor proteins. Proper spindle function requires that microtubule dynamics are stabilized at anaphase. This change in microtubule dynamics is dictated by a shift in the balance of kinase and phosphatase activities in favor of the phosphatases. The finding that cells simultaneously lacking the polo-like kinase Cdc5 and the phosphatase Cdc14 cannot progress through anaphase albeit having cleaved cohesin due to defects in spindle elongation, challenges the view of mitotic exit as a time for protein dephosphorylation. We identified the kinesin 5 motor protein Cin8 as a key target of the “Cdc14-Cdc5” spindle elongation pathway. We show that besides being dephosphorylated by Cdc14, Cin8 is also phosphorylated by Cdc5 on residues S409 and S441, and that this phosphorylation is crucial for the function of the kinesin in anaphase spindle elongation. The finding that Cin8 is simultaneously a substrate of a kinase and a phosphatase sheds light on the complexity of mitotic exit regulation. Since it appears that phosphorylation and dephosphorylation events are equally important to the point that kinases and phosphatases cooperate to regulate the same substrates, the view of mitotic exit as the realm of phosphatases is dismantled and the continuous need for single molecule studies in addition to global analyses investigation is put forward.

Roberta Pascolutti


Molecular characterization of the role of Cbl proteins in EGFR endocytosis
c-Cbl is the major E3 ligase involved in ubiquitination of Epidermal Growth Factor Receptor (EGFR). Ubiquitination by c-Cbl targets EGFR to lysosomal degradation. Its involvement at early internalization steps is still debated, due to the fact that multiple internalization pathways were described. Moreover, in addition to its role as an E3 ligase, c-Cbl works also as an adaptor, by recruiting several proteins involved in the early phases of clathrin-mediated endocytosis (CME). Importantly, c-Cbl has been found mutated in different disorders, from myeloproliferative disease to Noonan syndrome and non-small cell lung cancer (NSCLC). Most of these mutations are located within the Ring finger domain and in the regulatory linker region, therefore predicted to affect E3 ligase activity. In order to draw a more precise molecular picture of c-Cbl activity in EGFR ubiquitination and endocytosis, we investigated the effects of different set of cancer-relevant mutations, combining two distinct approaches: 1) RNA interference-based functional assays and 2) in vitro ubiquitination assays. 1) First, we characterized the effect of the knockdown of Cbl proteins on EGFR ubiquitination and endocytosis, confirming its essential role in NCE, by ubiquitinating the EGFR. Importantly, reconstitution experiments with RING finger mutants demonstrated that c-Cbl E3 ligase activity is also required for CME and is possibly exerted on endocytic adaptors. In agreement, Eps15 monoubiquitination is impaired upon c-Cbl KD. 2) We were able to reconstitute the EGFR ubiquitination reaction in vitro, and now we can use this tool to study the molecular details of c-Cbl catalysis.

Paola Pellanda


Structure-function analysis of Myc/Max-DNA binding
Myc oncoproteins are transcription factors of the basic-Helix-Loop-Helix Leucine-zipper (bHLH-LZ) family. As pre-requisite for DNA binding, Myc has to dimerize with the bHLH-LZ partner Max and together they bind preferentially to the E-box motif CACGTG. The ability of Myc to bind DNA in vivo, however, is not stringently regulated by the presence of the E-box, since many genomic sites targeted by Myc do not contain this motif. Hence, we still need to fully comprehend how Myc recognizes its genomic targets and to what extent sequence-specific DNA binding contributes to this process. Based on the crystal structure of the DNA-bound Myc/Max dimer, we generated a Myc mutant in which two residues engaged in sequence-specific contacts (H359 and E363) were mutated to Alanine (MycHEA), and compared this with a mutant in which three residues involved in DNA backbone interactions were mutated to Alanine (MycRA). While both mutants showed impaired E-box recognition in vitro, their over-expression in murine fibroblasts revealed very different genome-interaction profiles, MycRA showing no detectable DNA binding, and MycHEA retaining about half of the binding sites seen with Mycwt. The analysis of the binding intensity of Mycwt and MycHEA at their binding sites revealed that, while Mycwt bound more strongly the sites containing the E-box, MycHEA bound to DNA was sequence-independent, confirming that the mutant lost the sequence-specific recognition ability. In spite of this residual DNA-binding activity, MycHEA was profoundly impaired in its biological functions, undistinguishable from MycRA, strongly suggesting that the E-box recognition is essential for Myc’s biological function.

Vera Pendino


Role of miRNA in regulating the dosage of duplicated cancer genes
MicroRNAs (miRNAs) are endogenous short sequences that interfere with messenger RNAs leading to translational inhibition. The peculiarity of miRNA-mediated regulation resides in their ability to finely tune the levels of protein expression. From an evolutionary perspective, this miRNA property may play a role in buffering the perturbation introduced by gene duplication. In this thesis, I developed a strategy to date the evolutionary appearance and the origin of duplication of miRNA targets and studied the relationships between the evolution of miRNA regulatory system and the evolution of their human target genes. I found that singleton targets were mostly acquired with metazoans or later, while duplicated targets are old genes, whose orthologs can be traced back in early eukaryotes and in prokaryotes. Interestingly, also the expansion of these duplicated genes occurred in metazoans. Moreover, duplicated targets tend to be regulated by a higher number of miRNAs than singletons, and often these miRNAs are also duplicated. The results of my study strengthen the idea that miRNAs triggered the fate of genes after duplication, contributing to the acquisition of new genes and to the increase of morphological complexity during the evolution of metazoans. Further they indicate that miRNA regulation is conserved among paralogs, which are co-regulated by the same miRNAs. In the context of cancer, this study contributes a better understanding of the properties of cancer genes, since they undergo miRNA regulation more than the rest of human genes. Indeed, cancer genes are fragile toward perturbations and dosage modifications may affect their function.

Valentina Petrocelli


Role of Ezh2 methyltransferase activity in the maintenance of MYC-driven B cell lymphomas
The Polycomb group protein Ezh2 catalyzes the Histone H3 lysine-27 trimethylation (H3K27me3) within the Polycomb Repressive Complex 2 (PRC2). PRC2 exerts a critical control over the expression of a large set of target genes controlling important biological functions, including cell proliferation, differentiation and stem cell selfrenewal. Aberrant Ezh2 function is commonly observed in several cancer types and is due to deregulated enzymatic activity and/or expression of the Polycomb protein. Studies in preclinical models have started to reveal the importance of Ezh2 in B cell lymphomagenesis. In contrast, little is known about the effects of Ezh2 deregulated function/constitutive expression in B cell tumor maintenance and progression. The present study addresses this issue taking advantage of a MYC-driven mouse lymphoma model, featuring high Ezh2 expression as a result of malignant B cell transformation. Conditional, genetic inactivation of Ezh2 methyltransferase activity in aggressive primary Burkitt-like mouse B cell lymphomas led to the identification of two classes of tumors, differentially responding to the loss of Polycomb function. In type-1 lymphomas, Ezh2 inactivation impaired clonal tumor growth starting from single lymphoma cells. Instead, type-2 lymphomas were largely resistant to the loss of Ezh2 catalytic function, giving rise to a substantial number of Ezh2 mutant clones. Transcriptome analyses allowed the identification of a molecular signature discriminating type-1 from type-2 lymphomas, including genes controlling cell cycle progression, DNA replication and cell survival, which were more expressed in type-2 tumors. These results correlated with a more aggressive behavior of type-2 lymphomas when transplantated into immunoproficient hosts. The growth of rare Ezh2 mutant subclones, established from type-1 lymphomas, was impaired by the treatment with an Ezh1/2 small molecule inhibitor, identifying the Ezh2 paralog, Ezh1, as a determinant of resistance of tumor cells to Ezh2 inactivation. Ezh2 inhibition led to genome wide loss of H3K27me3, which was comparable between lymphoma types. However, while the loss of H3K27me3 at target genes in type-1 lymphomas failed to alter their expression, in type-2 lymphomas Ezh2 targets were in most cases deregulated following the loss of the histone mark. Based on these results, we propose that Ezh2 mutant subclones from type-1 lymphomas select an H3K27me3-independent mechanism to ensure correct regulation of Ezh2 target genes, which is needed for tumor growth. We also find that residual H3K27me3 is deposited at the promoter of new genes by a non-canonical PRC2/Ezh1, in Ezh2 mutant subclones from type-1 lymphomas. This activity alters the expression of target genes contributing to tumor growth. We finally report the isolation of clonal variants from type-1 lymphomas that acquire secondary resistance to pharmacological Ezh1/2 inhibition. The latter tumors (together with type-2 lymphomas) will be instrumental to unravel the genetic bases of resistance of MYC-driven lymphomas to PRC2 inhibition. Anti-Ezh2 inhibitors are currently being tested in phase-1 and -2 clinical trials for the treatment of both solid and blood cancers including B cell lymphomas. Our studies highlight the importance of understanding in more detail the mechanisms of action of Ezh2/PRC2 in tumors, in order to identify those that may benefit from anti-Ezh2 therapies. Our results also provide evidence for mechanisms of lymphoma resistance to Ezh2 inhibition and suggest strategies to circumvent such resistance.

Andrea Piunti


Polycomb role in cellular proliferation and transformation
PRC1 and PRC2 regulate several cellular functions among which the ability to promote proliferation is a main feature that links Polycomb (PcG) activity to cancer development. PcGs are involved in silencing the tumour suppressor locus Ink4a-Arf whose products positively regulate pRb and p53 functions. Enhanced PcGs activity is a frequent feature of human tumours and PcGs inhibition has been proposed as a strategy for cancer treatment. However, the recurrent inactivation of pRB/p53 responses in human cancers opens a question regarding the ability of PcG proteins to control cellular proliferation independently from the Ink4a/Arf-pRb-p53 axis. Here we demonstrate that PRC1 and PRC2 control cellular proliferation and tumor development independently of the Ink4a-Arf, pRb and p53 pathways. We provide evidences that the PRCs regulate DNA replication and show that PcG loss of function induces severe defects in replication fork progression and symmetry. Collectively, these data demonstrate that PRC1 and PRC2 can act independently of major cell cycle restriction checkpoints and identify DNA replication as a novel activity by which PcGs can regulate cell proliferation and tumor development.

Sara Polletti


Reorganization of the macrophage epigenome during sustained stimulation
The chromatin of cells whose main function is to sense and react to environmental inputs, such as macrophages and other innate immune cells, undergoes rapid modifications in response to microenvironmental signals and provides general paradigms of how epigenomes are dynamically reorganized in a changing environment. A short exposure of macrophages to endotoxin (lipopolysaccharide, LPS) strongly activates transcription of hundreds of inflammatory genes. Conversely, a sustained stimulation results in a state of hypo-responsiveness to a subsequent microbial stimulation, which is commonly referred to as endotoxin tolerance. We used nascent RNA-seq and ChIP-seq to characterize genes and cis-regulatory regions that are differentially activated in unperturbed, LPS-stimulated and LPS-tolerized primary mouse bone marrow-derived macrophages (BMDM). We characterized promoters and enhancers by mapping the methylation and acetylation states of associated histones and we identified differentially expressed genes by nascent RNA profiling. We clustered genes into different subsets based on their activity profiles and a detailed analysis of these datasets allowed us to dissect the mechanisms underlying functional switches in the macrophage gene expression program during sustained inflammation. In particular, IRF7 played a key role in the transcriptional regulation of sustained genes as indicated by the impact of its depletion. Transient genes were mainly regulated by the Early Growth Response (EGR) and Nuclear factor-κB (NFkB) family transcription factors, which are both downstream effectors of the TLR4 signaling pathway. Biochemical analysis of the key players of this signaling cascade revealed an almost complete exhaustion of the pathway after sustained LPS treatment. These data suggest that the failure to reactivate those transcription factors that areinvolved in the transcriptional expression of transient genes is likely due to the hyporesponsive state of the TLR4 signaling pathway in cells exposed to a sustained LPS stimulation.

Mahshid Rahmat


Functional dissection of the histone demethylase Jmjd3 in B cell lymphopoiesis
Histone H3 lysine-27 trimethylation (H3K27me3) is an epigenetic mark that exerts a critical role in heritable gene repression. Modulation of H3K27me3 levels influence cell proliferation, survival and differentiation. In mammalian cells, the Jumonji-C containing proteins JMJD3/KDM6B and UTX/KDM6A are the only known enzymes involved in H3K27me3 demethylation. Jmjd3 has been previously shown to play an important role in mediating macrophage driven inflammatory responses and in in regulating somatic cell reprogramming and cellular senescence. To study the role of JMJD3 in B lymphocyte development and activation, I generated Jmjd3 conditional knock-out mice (JMJD3fl). Analysis of B cellspecific Jmjd3 KO mice revealed that Jmjd3 regulates the size of the B cell progenitor pool acting primarily on the pre-B cell compartment. Jmjd3 deficient animals showed also an increase in the fraction of splenic marginal zone B cells that was associated to a reduction in peritoneal cavity B-1a B cells. In vivo BrdU labeling assays suggested a longer lifespan of Jmjd3 mutant mature B cells, which was not dependent on improved cell survival. In vitro stimulation assays revealed a selective defect of Jmjd3 mutant B cells to proliferate in response to the TLR4 ligand LPS, which was alleviated by IL- 4 co-stimulation. Jmjd3 was critical to drive the first one-to-two cell divisions following LPS stimulation suggesting a critical role in the initial activation of the resting B cells. RNA sequencing data, revealed a comprehensive control exerted by Jmjd3 on the expression of a substantial number of cell-cycle regulated genes including those cyclins, CDK inhibitors and factors involved in DNA replication and mitosis. Regulation of gene expression mediated by Jmjd3 was not associated with measurable changes in global H3K27me3 levels. All together these results identify Jmjd3 as an important regulator of B cell lymphopoiesis and a selective effector of B cell innate immune responses.

Alexandra Raileanu


An Electrochemical Detection System for Thiocholine Using Cluster-Assembled Zirconia Modified Gold Electrodes
Rapid, low cost, sensitive and user friendly detection methods for pesticides that contain organophosphate compounds are important for environmental pollution control. Even though the current detection methods are very precise, they are meant only for laboratory use, they require highly trained personnel, are expensive, cumbersome and the usage of a large number of reactants and chemicals is needed. Nanotechnology combined with specific enzymes that are influenced by the presence of organophosphates could offer the alternative of designing specific biosensors that are more suitable for on-site detection. With this PhD work I intend to approach one of the drawbacks that are commonly found in designing a platform for portable electrochemical detection systems: a good communication between the materials that make up a transducer and the biological component which will interact with the analyte. Cluster-assembled zirconia nanostructured thin films, as a transducer component, have been optimized for a suitable immobilization of Acetylcholinesterase enzyme that acts as a catalyst for the breakdown of Acetylcholine. Moreover, the implementation of zirconia has shown to improve the electron transfer when used as part of the transducer in electrochemical assessments, which makes it a suitable candidate for the development of detection platforms for environment monitoring based on electrochemical techniques.

Emanuele Ratti


The context of discovery of data driven biology
My PhD dissertation aims (1) at reconstructing the structure of the context of discovery of ‘data-driven’ (big data, data intensive) biology and (2) at comparing it to traditional molecular approaches. Within the current debate in philosophy of science, ‘traditional approaches’ in molecular biology should be understood as the discovery and heuristics strategies identified by mechanistic philosophers such as Carl Craver and Lindley Darden. Therefore, key questions of my thesis are: what is the structure of discovery of datadriven biology? Is data-driven biology methodology different from traditional molecular approaches?  The reason for doing such an analysis comes from a recent controversy among biologists. In particular, sides disagree on whether high throughput sequencing technologies are stimulating the development of a new scientific method somehow irreducible to traditional approaches. I will try to disentangle the debate by reconstructing and comparing data-driven and traditional methodologies. The dissertation is composed of five chapters. The first chapter deals with methodological issues. How do I compare data-driven and traditional molecular biology structures of discovery? Mechanistic philosophers have extensively characterized the discovery structure of traditional molecular biology. However, there is not such an analysis for data-driven biology. In order to do this, I will critically revise the discovery/justification distinction. The debate on discovery/justification has provided valuable tools on how discovery strategies might be conceived, and it is clearly one of the main forefathers of recent philosophical discussions on scientific methodologies in biology and physics. In Chapter 2 I shall to try to infer a full-fledged account of discovery for datadriven biology by means of the philosophical tools developed in Chapter 1. This analysis will be done in parallel to the investigation of key examples of data-driven biology, namely genome-wide association studies and cancer genomics. In Chapter 3 I analyze the epistemic strategies enabled by biological databases in data-driven biology. In Chapter 4, I will show how the discovery structure of ‘traditional molecular biology’ can be more efficiently rephrased through the same theoretical framework that I use to characterize data-driven biology. Since data-driven and traditional molecular biology seem to adopt the same discovery structure, one might consider the controversy motivating my research ill posed. However, in Chapter 5 I shall argue that there is still a valuable reason of disagreement between the sides. Actually, data-driven and traditional molecular biology endorse different cognitive values, which provide the criteria for evaluating models and findings as adequate or not. Here one might say that, although the structures of discovery (i.e. how  reasoning and experimental strategies are structured and depend on each other) of the two sides are the same, the contexts of discovery  (i.e. the set of both reasoning/experimental strategies and  epistemic values/background assumptions that motivate discovery) are different. Therefore, in this last chapter I shall pinpoint the cognitive values behind traditional and data-driven biology, and how these commitments stimulate the heated disagreement motivating my research.  

Micol Ravà


Functional dissection of ST18 in liver cancer
The molecular mechanisms and pathways responsible for the progression of hepatocellular carcinoma (HCC) remain to be fully characterized. Among the genetic lesions associated with HCC progression, Shukla et al. (2013) identified insertions of the L1 transposon proximal to the gene encoding the zinc-finger DNA-binding protein ST18 (suppression of tumorigenicity 18) and suggested that this actually functions as an oncogene in HCC. However, functional evidence for a cancer-promoting activity of ST18 and insight into its mode of action are missing. Here, I pursued the functional characterization of ST18 in a mouse model of HCC based on ex vivo transformation and subcutaneous transplantation of embryonic hepatoblasts. ST18 was undetectable in either normal liver or cultured hepatoblasts, but was induced in the subcutaneous tumors. ST18 was also expressed in either chronically or acutely inflamed mouse livers (as assessed in Mdr2-/- or LPS-treated mice) as well as in human Progressive Familial Intrahepatic Cholestasis 2 (PFIC2: a condition associated with chronic inflammation), suggesting its induction by inflammatory stimuli. The knockdown of ST18 delayed tumor formation or, if induced in already formed tumors, led to rapid hemorrhage, pervasive morphological changes in the tumor cells reminiscent of an epithelial-to-mesenchymal transition (EMT) and eventually tumor regression. RNA profiling revealed that ST18 silencing caused expression of EMT-associated genes, among others. Previous studies have linked inflammation to the induction of EMT in other epithelia: we hypothesize that the concomitant activation of ST18 constitutes a safeguard against EMT, inactivation of this control mechanism causing the dramatic phenotypic switch observed in our model. These data warrant further evaluation of the mode of action of ST18 and of its potential value as a therapeutic target in HCC.

Francesca Reggiani


GM-CSF and MMP9 are key regulators of the effect of adipose progenitors over breast cancer onset and metastatic progression
Recent epidemiological and clinical data underlined the critical role of obesity in breast cancer (BC) progression. A population with progenitor-like phenotype (CD45-CD34+) was detected in white adipose tissue (WAT) and was reported to support local and metastatic BC. This population is composed by distinct WAT progenitors: adipose-derived stem cells (ASCs) and endothelial progenitor cells (EPCs), displaying complementary role in BC progression. The molecular mechanisms involved in this interaction need to be clarified. Two proteins were significantly up-regulated in WAT-derived progenitors after being co-cultured with several BC cells: Granulocyte-macrophage colony-stimulating factor (GM-CSF) and Matrix metallopeptidase 9 (MMP9). Both factors were detected over-expressed in xenograft models, when co-injected with BC and human WAT progenitors. ASC and EPCs displayed similar ability to induce GM-CSF/MMP9, suggesting a complementary role in their release. The neutralization of GM-CSF in diet-induced obese (DIO) syngeneic mice led to reduced intratumor vascularization and strong impairment of immunosuppressive microenvironment. This resulted in a significant impairment of local BC growth and a slower metastatic progression. Conversely, MMP9 inhibition reduced neoplastic angiogenesis and significantly decreased local and metastatic tumor growth. Metformin was reported to significantly affect tumor progression and neoplastic angiogenesis, targeting both BC and WAT cells. In the present study, Metformin inhibited GM-CSF and MMP9 release from WAT progenitors in vitro and in xenograft models. Collectively, these results indicate GM-CSF and MMP9 as new potential targets to prevent the pro-tumorigenic effect of WAT progenitors on BC. Furthermore, Metformin ability to reduce GM-CSF and MMP9 supports Metformin administration in clinical studies.

Catarina Remedios


Nanoparticles for mucosal vaccine delivery 
Currently, injection-based vaccination is the most common method for influenza immunization. However, parenteral vaccination fails to induce mucosal immune responses, representing an important first line of defense. Under the framework of the UniVacFlu project, this study was undertaken to assess the potential of the universal influenza vaccine candidate CTA1-3M2e-DD as an oral vaccine, along with that of cationic polysaccharide nanoparticles (NPL) as an oral vaccine delivery system. We found that, while CTA1-3M2e-DD revealed a poor ability to cross the intestinal epithelium and target lamina propria antigen-presenting cells, NPL readily overcome the intestinal barrier and are phagocytosed by both CX3CR1+ macrophages and CD103+ dendritic cells. Two different routes of NPL uptake were identified: one depends on Goblet cell-associated passages that allow the transfer of high amounts of NPL from the lumen into the intestinal lamina propria; the second relies on the direct acquisition of NPL by CX3CR1+ cells in Peyer’s patches through extension of trans-epithelial dendrites. Moreover, using our model of oral immunization that protects NPL from the stomach pH, the vaccine vector was able to deliver the loaded antigen in the intestinal lamina propria and enhanced antigen presentation to CD4+ T lymphocytes in different organs. Despite increasing antigen presentation, NPL did not induce tolerance towards the formulated antigen and a Th1 immune response was found at the level of the Peyer’s patches. We also identified the contribution of the starvation period in the immune response induced by the NPL formulation in our model of oral immunization.

Silvia Restelli


Regulatory mechanisms implicated in the control of Numb asymmetric partitioning at mitosis of adult mammary stem cells
The cell fate determinant Numb has been recently shown to control the asymmetric outcome of self-renewing divisions in the normal mammary stem cell (MaSC) compartment. Loss of Numb results in the appearance of stem cells (SCs) endowed with an enhanced symmetric self-renewal ability (leading to expansion of the SC compartment) and with tumorigenic potential. The role of Numb in MaSCs is linked to its ability to asymmetrically partition at mitosis into the progeny that retains the SC identity, where it sustains the tumour suppressor activity of p53. Phosphorylation of Numb by the Par Complex (Par3/Par6/atypical PKC) is an evolutionary conserved mechanism to specify Numb asymmetric partitioning at SC mitosis. In this thesis work, we used a molecular genetics approach to demonstrate that a proper cycle of Numb phosphorylation and de-phosphorylation is essential to guarantee the asymmetric partitioning of Numb at mitosis of MaSCs. We also propose a conserved role for the atypical PKCx isoenzyme and Par3 in the regulation of these events. Moreover, our preliminary data show that Numb phosphorylation mediated by aPKCx impairs Numb interaction with p53, suggesting that the regulation of Numb phosphorylation can be also linked to its tumour suppressor function.

Abrar Rizvi


Role of IRSp53 in epithelial polarity establishment and lumen morphogenesis
Fused Filament Fabrication (FFF) three-dimensional printing have attracted much attention for fabrication of microfluidic platforms used to construct electrochemical microfluidic biosensors because of high process speed, low production costs and the possibility of manufacturing directly from virtual data. Because of poor adhesion between metal electrodes fabricated using conventional techniques and FFF printed thermoplastic substrates, electrodes are usually integrated into the devices either modularly or using adhesive layers placed at the bottom of fluidic channels. These have hindered the exploitation of FFF for scale-up manufacturing of monolithically integrated microfluidic biosensors. In this work, supersonic cluster beam deposition (SCBD) was employed to fabricate strongly anchored nanostructured electrodes integrated into FFF printed microfluidics platforms. SCBD enables the formation of well-adhering metallic thin film electrodes by implanting supersonically accelerated neutral metal clusters into polymeric substrates. The SCBD also enables deposition over large areas using noble metals and metal oxides with precisely controlled geometry and surface topography. A novel integrated manufacturing approach was developed and optimized to couple SCBD fabricated electrodes with consumer-grade FFF printed microfluidics, employing acrylonitrile butadiene styrene as the base material, to develop a three electrodes configuration electrochemical sensor on-a-chip. Electrochemical investigation performed using stagnant ferro/ferricyanide probe showed that the integrated device possesses high sensitivity and functionality as an electrochemical sensor. In addition, in-channel laminar flow electrochemical detection conducted using the same probe showed robust stability in the system response for online dynamic detection. The integrated platform could be employed for various customized clinical, industrial, and environmental sensing applications.

Michela Roccuzzo


Regulation of chromosome segregation by conserved phosphatase Cdc14 and kinase Cdc5
The faithful transmission of the replicated genome from the mother to the daughter cell requires the correct establishment of linkages between the duplicated chromosomes (sister chromatids) and their bi-orientation on the mitotic spindle. Chromosome segregation initiates only after each sisters pair is correctly aligned onto the microtubules emanating from the spindle poles. Next, Esp1-mediated cleavage of cohesin is required to trigger anaphase onset while the physical segregation of the separated sisters is next driven by spindle activity. However, this scenario appears to be more complicated involving additional factors driving the sister chromatid segregation process (i.e. the Top2-mediated resolution of replication catenates). In budding yeast, anaphase progression and exit from mitosis require the protein phosphatase Cdc14 whose activation relies on two consecutive protein pathways, the FEAR network and the MEN. As the polo-like kinase Cdc5 is a component of both pathways its activity is essential to Cdc14 release and in its absence Cdc14 is never released. By combining loss-of-function alleles of Cdc5 and Cdc14 we obtained double mutant cells that had cohesin cleaved but still arrested with undivided nuclei and short bipolar spindles. Anaphase spindle elongation initiates quickly after cohesin removal (anaphase A) and then switches to a slower elongation rate (anaphase B) due to changes in spindle behavior mediated by motor proteins and microtubule-associated enzymes. Although some residual cohesion between sister chromatids seems to contribute to the terminal phenotype of cdc14 cdc5 cells, our data indicate that anaphase B is the main mitotic defect of these cells. We conclude that Cdc5 and Cdc14 are redundantly involved in activating spindle activity following cohesion resolution, suggesting the existence of a regulatory network that coordinates sister chromatid separation with spindle elongation after cohesin cleavage. Importantly, we identified the motor protein Cin8 as a (direct or indirect) target of Cdc5 in the regulation of spindle elongation.

Sara Rohban


Genetic dissection of the Myc-induced DNA Damage Response
In pre-cancerous lesions the overexpression of oncogenes such as Myc not only drives aberrant cellular proliferation, but also triggers a strong DNA damage response (DDR) that is in part due to DNA damage accumulating at the level of stalled replication forks. This oncogene-induced DDR is an effective barrier to cancer development and represents a relevant tumor suppressive mechanism. Conversely, at later stages of malignancy DDR signaling may function in favor of cancer progression. Such tumor promoting role of DDR may be needed for cancer cells to avoid accumulation of cytotoxic DNA damage under high level of oncogene-induced replication stress. In an effort to identify the modulators of Myc-induced replicative stress, we carried out a high-throughput RNAi screen based on immunofluorescence detection of ϒH2AX, a DNA damage marker. Quantification of the number of cells and the percentage of ϒH2AX-positive cells, identified hits that exhibited differential cell viability and/or enhanced ϒH2AX signal in Myc-overexpressing cells compared to normal cells. Validated hits encompass a variety of pathways and biological processes and have different molecular functions. For further mechanistic investigations we selected Rad21, a component of the cohesin complex. We provide evidence that while Rad21 is necessary for proper and efficient DNA synthesis, replication reinforcement imposed by Myc in Rad21-depleted cells results in replicative stress. In addition, we observed that Myc, as a transcription factor, could partially rescue transcriptional alterations due to Rad21 depletion. The conflicts between DNA replication and transcription in Rad21-depleted cells upon Myc activation may be the source of increased R-loops detected in these cells. In summary, by means of a genetic loss of function screen we identified several candidates that may be involved in protecting Myc-overexpressing cells against ample replicative stress, thus revealing targets for potential therapeutic intervention in Myc-driven cancers.

Dalia Rosano


Unravelling the molecular mechanisms underlying oncogene induced DNA damage
Activation of oncogenes in normal cells induces DNA damage accumulation and evokes a DNA damage response (DDR) that has been proposed to behave as a barrier against malignant transformation by inducing either apoptosis or cellular senescence, and therefore avoiding the clonal propagation of the transformed phenotype. Two main mechanisms have been proposed to explain the role of oncogenes in inducing DNA damage: replication stress and oxidative stress. However, further investigations are needed to fully decipher the direct mechanistic link between oncogene activation and DNA damage accumulation. To investigate the molecular mechanisms underlying PML-RARαinduced DNA damage, we localized on a genome-wide scale the DNA lesions induced by the oncofusion protein by performing ChIP-sequencing analyses of γH2AX DNA damage marker. Moreover, to understand the nature of PML RARαinduced DNA damage, it was compared to the DNA damage induced by irradiation or by the endonuclease activity of a restriction enzyme. We found that PML-RARα induces site-specific γH2AX accumulation at its binding sites, where it deregulates the expression of its target genes. Indeed, PML-RARα direct targets were associated with the largest and most intense γH2AX domains, as a consequence of recurrent DNA lesions. Therefore transcription deregulation and/or its interference with replication might be responsible for DNA damage accumulation at PML-RARα binding sites.

Eleonora Rossi


An innovative strategy for adipose tissue reconstruction
Despite clinical treatments for adipose tissue defects have certain grades of efficacy, in particular for breast tissue reconstruction, many drawbacks are still affecting the long-term survival of new-formed fat tissue. To overcome this problem, numerous scaffolding materials based on either synthetic or natural polymers have been investigated. In parallel, devitalized adipose tissue has shown great in vivo potential to enhance regeneration and instruct repair of damaged soft tissues. Our aim was to combine the positive features of a structural synthetic polymer and of an engineered, devitalized extracellular matrix (ECM) to obtain a hybrid construct for the treatment of adipose tissue defects. Towards achieving this goal, we adopted a biologically and mechanically driven design to fabricate an RGD-mimetic poly(amidoamine) oligomer macroporous foam (OPAAF) for adipose tissue reconstruction. The OPAAF is characterized by an interconnected porous network and mechanical properties resembling the native adipose tissue. Moreover, OPAAF supported cell adhesion, proliferation and adipogenesis in vitro together with adipose tissue infiltration in vivo. Afterwards, we further implemented the adipoinductive potential of the OPAAF by decoration with a devitalized adipose tissue matrix deposited by human Adipose Stromal Cells (hASCs). The hybrid ECM-OPAAF showed to have an adipoinductive effect on hASCs in absence of any growth factor in vitro and promoted adipogenesis in vivo. Overall, these results proved that our approach can provide an alternative strategy for adipose tissue reconstruction based on the use of patients cells for the generation of custom made hybrid scaffolds.

Alessandra Rossi


Role of the Polycomb group proteins in the adult intestinal stem cells homeostasis
Polycomb group proteins (PcG) are among the most important gatekeepers that ensure the correct establishment and maintenance of cellular identity in metazoans. This occurs by modifying chromatin through the activity of two Polycomb Repressive Complexes (PRC1 and PRC2) that deposit H2A ubiquitylation and H3K27 methylation respectively, in order to guarantee repression of their target genes. Although the development of PRC2 inhibitory compounds is becoming a very promising strategy for specific cancer treatment, the controversial role of PcG proteins, acting as oncogenes or tumor suppressors in a tissue/cancer specific manner, prompt us to further investigate the role PcG proteins in regulating adult tissue homeostasis. Using different genetic models, we have found that PRC1 activity is required for the integrity of the mouse intestinal epithelia. More in detail, PRC1 activity is required for the self-renewal of the intestinal stem cells (ISCs) via a cell-autonomous mechanism that is independent of Ink4a-Arf expression. Using high-throughput transcription and location analysis, we have dissected the direct transcriptional pathways regulated by PRC1 in ISC showing that PRC1 inactivation induces a loss of ISC identity as a result of a massive up-regulation of non-lineage specific transcription factors that can directly inhibit the transcriptional activity of the ß-Catenin/Tcf4 complex. Overall, we propose that PRC1 control the self-renewal of ISC by positively sustaining Wnt transcriptional activity also in the presence of oncogenic mutations that constitutively activate the Wnt pathway in intestinal tumors.

Silvia Rossi


Interplay between the DNA helicases Pif1 and Rrm3, the nuclease Dna2 and the checkpoint pathways in the maintenance of the DNA replication fork integrity
The DNA damage response is a pathway responsible for the maintenance of genome integrity. In my thesis I focused on the investigation of the modulation of ATM activity, a DNA damage response master kinase, by NOTCH1 receptor. Here I show that NOTCH1 inhibits DNA damage response activation. This inhibitory effect of NOTCH1 is not mediated by its transcriptional activity, but it is the result of direct binding between NOTCH1 and ATM kinase. I show that NOTCH1 binds to the FATC domain of ATM, and this results in an inhibition of ATM kinase activity. Furthermore, I provide evidence that NOTCH1-mediated ATM inhibition does not result from the impairment of ATM recruitment to DNA double-strand breaks. Rather, I show that NOTCH1 competes with FOXO3a transcription factor for the binding to the FATC domain of ATM and that over-expression of FOXO3a prevents NOTCH1-mediated ATM inhibition. As the exact function of FOXO3a in ATM activation was unclear, I sought to understand molecular mechanisms underlying NOTCH1-mediated ATM inactivation and the role of FOXO3a as an opposing factor in this process. I discovered that FOXO3a forms a direct complex with KAT5 lysine acetyl transferase that is critical for ATM activation upon DNA damage. Moreover, I observed that FOXO3a was necessary for the formation of a complex between ATM and KAT5. Surprisingly, I observed that NOTCH1 was not only impairing ATM-KAT5 interaction, but also FOXO3a-KAT5 one. This unexpected observation led me to the discovery that FOXO3a-KAT5 interaction is restricted to the formation of this three-protein complex together with the ATM kinase. Next, I demonstrated that induction of FOXO3a nuclear localization as well as inhibition of NOTCH1 increases ATM activation in NOTCH1-driven cancer cells, which leads to augmented DNA damage-induced cell death. Finally, I show that, in addition to ATM, NOTCH1 interacts also with other PI3K-like kinases: DNA-PKcs and ATR. Although I did not observe a significant impact of NOTCH1 on ATR kinase activation in the experimental settings I used, I observed an impaired activation of DNA-PKcs, which however did not result in a significant reduction of DNA damage repair in NOTCH1-expressing cells.  

Pierluigi Rossi


Role of the histone demethylase LSD1 in the regulation of differentiation of acute promyelocytic leukemia cells
Acute promyelocytic leukemia (APL) is characterized by the appearance of blasts, insensitive to physiological retinoic acid (RA)-dependent differentiation and in the majority of the cases, expressing PML-RAR fusion protein. PML-RAR works as a dominant negative counterpart of retinoic acid receptor (RAR) and represses its target genes by cooperating with many epigenetic regulators. Some of the known epigenetic enzymes involved in the APL pathogenesis also interact with LSD1 in other systems. LSD1 is a demethylase involved in transcriptional regulation, mainly acting on dimethylated lysine 4 of Histone H3 (H3K4me2). We showed that LSD1 inhibition sensitizes an APL cell line to physiological RA concentration. Here we characterize for the first time mechanistic insights of LSD1 activity in PML-RAR expressing APL cells. We determined the genomic distribution of LSD1, in particular LSD1 binds both promoters and candidate enhancer regions. We found that LSD1 works as finetuners of genes involved in differentiation and cell growth control as assessed by gene ontology analysis. Moreover, LSD1 modulates H3K4me2 levels at regions enriched in binding sites of master regulators of the myeloid/monocytic lineage, suggesting their regulatory potential. We also described a previously unknown large fraction of genomic loci bound by LSD1 and PR. Commonly bound regions show a peculiar H3K4me2 enrichment and a subset of them resulted dynamically regulated upon differentiation, suggesting a functional interplay between the two proteins in reshaping the local chromatin environment. Overall, our findings contribute to the mechanistic understanding of the role of LSD1 in the sensitization of APL cells to differentiation.

Francesca Rossiello


Persistent DNA damage at telomeres, caused by trf2-mediated DNA repair inhibition, Triggers cellular senescence and is associated with primates ageing
The DNA damage response (DDR) coordinates DNA repair events and transiently arrests cell-cycle progression until DNA damage has been removed. If the damage is not resolved, cells can enter an irreversible cell cycle arrest called cellular senescence. In irradiation-induced senescent cells a large fraction of persistent DDR markers are associated with telomeric DNA, both in cultured cells and in in vivo tissues. The aim of my PhD project was to investigate the mechanism underlying this phenomenon. I showed that persistent DDR activation has a causative role for the senescence-associated cell cycle exit and that a double-strand break (DSB) within telomeric repeats is inducing a more protracted DDR activation compared with a nontelomeric one in human cells. The DDR persistency at telomeres is neither dependent on their heterochromatic state nor on TRF2 loss from telomeres during senescence establishment. Rather, TRF2 recruitment next to a DSB, in the absence of telomeric DNA, is sufficient to induce a more protracted site-specific DDR focus and to impair DSB repair in mouse cells. Ageing is associated with accumulation of markers of DDR activation. In terminally differentiated brain neurons from old primates, I observed DDR activation at telomeres that were not critically short. Taken together, these results strongly suggest that TRF2 inhibits DNA repair at broken telomeres, contributing to the accumulation of unrepaired, DDR-positive telomeres during ageing. This can in turn trigger cellular senescence and impair tissue homeostasis providing a mechanism for ageing also in non-proliferating tissues.

Anna Russo


Diet-specific epigenetic signature revealed by H3K4me3 and H3K27me3 data analysis in C57BL6 mice
Increasing evidences demonstrate that adapting to different environmental conditions is mediated by epigenetic changes, which can participate in cellular processes. In particular, adaptation to different caloric intakes is crucial for the organism’s fitness. Moreover, the phenotypic remodeling induced by different diets determines the susceptibility to life-threatening diseases: western diet is responsible for about 30- 35% of cancer cases, in addition to increased incidence of type 2 diabetes and cardiovascular diseases; while, caloric restriction has been shown to be the most powerful way to prolong lifespan and reduce cancer incidence in different experimental models. Based on the hypothesis that epigenetic changes represent the mechanistic link between diet and disease risk, the aim of this work is to investigate chromatin modifications induced by different diets in murine models to identify specific epigenetic profiles associated with different caloric intakes. Eight weeks old C57Bl/6 mice were divided in three groups and fed for 10 months with different diets: Standard Diet, Calorie Restriction without malnutrition, High Fat Diet. Livers were extracted and investigated by chromatin immunoprecipitation (anti- H3K4me3, anti-H3K27me3) and transcriptomic approach. Data analysis demonstrated that specific epigenetic profiles were associated to different diets. In particular, the distribution and frequency of H3K4me3 enabled the clustering of samples by diet-group. Moreover, functional annotation of genes showing an increased signal of H3K4me3 for HF or CR respect to SD on their promoter regions, resulted in significantly enriched “Type II diabetes”, for which obesity represents a critical risk factor, and “Circadian Rhythm” pathways, whose known to affect longevity.

Simone Sabbioni


Characterization of the molecular mechanism responsible for the loss of the tumor suppressor Numb in breast cancer
he tumor suppressor protein, Numb, safeguards against the emergence and the expansion of cancer stem cells (CSCs). Indeed, Numb loss occurs in ~30% of human breast cancers (BC) and correlates with biological aggressiveness and poor prognosis. A well-established mechanism responsible for Numb loss in BC is its aberrant proteasomal degradation. Thus, the identification of the proteasomal machinery responsible for Numb loss would not only provide clues as to the underlying lesions in Numb-deficient cancers, but could pave the way to the development of novel strategies to restore Numb expression and curbs tumorigenic potential of Numb-deficient BCs through the selective targeting of CSCs ensues. Through a siRNA-based high-throughput screening in Numb-deficient MDA-MB-361 cells, RBX1 and FBXW8, both components of the Cullin-RING E3-ligase complexes (CRLs), were identified as negative regulators of Numb stability in Numb-deficient cells. By comparing the Numb-deficient with the Numb-proficient (MDA-MB-231) BC cell models, we studied, through a combination of in vitro and in vivo pharmacological and genetic approaches, the effects on Numb expression levels and tumor phenotypes achieved upon the impairment of RBX1/FBXW8-CRL function. Overall, we formally demonstrated the involvement of a CRL complex in the excessive degradation of Numb in Numb-deficient BCs. Interestingly, pre-clinical in vivo models also provided proof-of- principle of the selective therapeutic value of targeting RBX1 and FBXW8 CRL components in these tumors. Moreover, our results supported the notion that the proteasome-inhibitor Bortezomib and the CRL-inhibitor Pevonedistat, two potential drugs already available in clinical context, could eventually be repositioned towards the treatment of Numb-deficient BCs.

Sara Samadi Shams


Identifying the molecular players associated with transition between pluripotent and totipotent-like state
The presence of rare transient cells within pluripotent stem cell population that resemble functional and transcriptional features of totipotent 2-cell-stage embryos has emerged several questions regarding their mechanism of reactivation. Although few recent works have aimed to characterize them both at transcriptional and functional level, the mechanism of transition to 2C-like state is poorly investigated. Here we identified a new pathway, whose activation through several compounds could induce transition to 2C-like state. Interestingly, inhibition of this pathway by specific inhibitors could collectively restrain the majority of the 2C-like state transcriptional alterations including MuERV-L and Zscan4 gene family. Finally, we found that developmental potential of induced 2C-like cells could be extended to embryonic plus extra-embryonic tissues in contrast to embryonic stem cells in culture. Our finding provides a better understanding of cellular plasticity at early embryonic state, which could be important for the potential therapeutic avenues.

Virginia Sanchini


Facilitate deliberation. Towards a redefinition of the bioethical expert in the public arena
In an increasingly specialized world, where the production of knowledge and its acquisition have become a collective enterprise, nobody can master all the fields alone. This has led to the proliferation of a myriad of experts, each of whom is specialized in a precise domain or subdomain. Given this picture, it is not surprising that people with training in bioethics are often referred to as ‘bioethics experts’ and/or ‘bioethical experts’. However, far from being confined to a specific social dimension, in today’s society bioethicists are a commonplace presence in an ever-increasing range of domains. In recognition of this phenomenon, the aim of this thesis is to explore the so-called issue of bioethical expertise. This means first and foremost exploring two main questions: whether an expertise in the field of bioethics might actually exist and what are, if any, the skills that the bioethical expert is endowed with, in order to understand whether this figure is on a par with professionals of other disciplines. After presenting a general review of the current literature on this issue and briefly indicating the different research paths this research topic might lead to, the focus will be narrowed to dwell on the main research questions this work aims to address: is there a legitimate role for the bioethical expert within the public arena, and, if so, what is this role? Far from being a straightforward question, this issue is further complicated by a vast literature decrying the threat posed by bioethicists with respect to the broader context in which they operate: liberal democracies. In taking these considerations seriously, this dissertation aims to ‘solve the paradox’ connected with the figure of bioethical expert and to propose a normative model of bioethical expert compatible with the tenets of (deliberative) democratic western democracies and, in particular, their public decision-making processes. This figure will be defined here as the facilitator of deliberation, thus recognizing the debt it has towards the political theories of deliberative democracy. This work is structured as follows: first of all the theoretical current debate concerning the topic of bioethical expertise is presented. Secondly, by narrowing the focus of investigation I concentrate on the domain in which the bioethical expert proposed here has to operate: public bioethics. The normative proposal of bioethical expert as facilitator of deliberation is then presented, remodelling the former in relation to the already existing figures occupying the public arena. Finally, the results of a preliminary experiment, designed to empirically explore this new figure, will be presented and discussed.

Francesco Santaniello


Changes of Replication Timing induced by PML-RARα
DNA replication is a cellular process that, starting from precise genomic loci, ensures the loyal and faithful inheritance, from one parental cell to each daughter cell, of the genetic instructions contained in the double-strand DNA molecule. Due to the complexity and the crucial importance of the DNA replication, this process must be tightly regulated in both space and time. Up to now, however, the time-related features of DNA replication, together with the factors that might impact the temporal dimension of this system, are yet poorly studied and described. Given the lack of standard methods able to recognize differences in Replication Timing, we developed an innovative bioinformatic method (DART; Differential Analysis of Replication Timing) to accomplish this task. Moreover, the application of this procedure to our Repli-seq data was instrumental to investigate whether PML-RARα may fulfil its tumorigenic potential by eliciting an alteration of the normal replication timing pace in cells. As a result, we found that, after its expression, PML-RARα indeed exerts a deregulative effect on Replication Timing, inducing some regions to replicate earlier (LtoE-shifted) and some other later (EtoL-shifted), with respect to control cells. We observed a close association between these differentially replicated regions and both pre-existing, and PML-RARα-related, transcriptional status and chromatin structure. Regions presenting a EtoL-shifted replication coincide with ‘active’ chromatin foci enriched for direct down-regulated targets of PML-RAR; at the opposite, regions with a LtoE-shifted Replication Timing show moderate epigenetic ‘active’ features and are enriched for indirect up-regulated targets of PML-RAR

Andrea Scelfo


Novel functions of Polycomb proteins and Ogt glycosyltransferase in chromatin regulation and transcriptional control
In spite of the wide knowledge about PRC2-dependent trimethylation of Lysine 27 of histone H3 (H3K27me3), the other forms of methylation on H3K27, namely mono (me1) and di-methylation (me2), are still poorly characterized. Using mouse embryonic stem cells (mESC) as model system, we were able to provide an extensive characterization of the functional properties of these methylation forms, highlighting their differential deposition along the genome, their fundamental role in the mechanisms of transcriptional regulation in mESC, and their potential implications during differentiation program. Our data demonstrated that while H3K27me1 was required for efficient transcription of genes and positively correlated with the deposition of H3K36me3, H3K27me2 was broadly deposited and protects the genome from aberrant firing of non specific cell type enhancers. The second project focused on the activity of the O-linked glycosyltransferase Ogt which is the only enzyme capable to catalyze O-linked GlcNAcylation within the cell. Sxc protein, the Drosophila orthologue of mammalian Ogt, is essential for Polycomb (PcG) function. Sxc null embryos showed lethal phenotypes like those observed in PcG null embryos. We wanted to investigate possible roles for Ogt in the context of PcG recruitment to chromatin and transcriptional regulation in mESC. We identified a novel partnership between Ogt and TET proteins which catalyze hydroxylation of methylated cytosine. We have shown that Tet1 protein recruits Ogt to target genes in proximity of transcription start sites (TSS) rich in cytosine-guanine dinucleotides (CpG). Tet1-Ogt co-localization at target genes correlated with low levels of cytosine modification, suggesting a role in the regulation of CpG island (CpGi) methylation.

Irene Schiano Lomoriello


Investigation of the role of the endocytic protein Epsin3 in EMT and mammary stem cell regulation in breast cancer
Endocytosis is a critical regulator of many cellular processes, including proliferation, migration, epithelial-to-mesenchymal transition (EMT) and maintenance of the stem cell (SC) compartment. Subversion of endocytic routes plays an important role in cancer development, including breast cancer (BC). We recently uncovered an oncogenic role for the endocytic protein Epsin3 (Epn3) in BC. By screening a large cohort of BC patients, we found that the EPN3 gene is amplified in ∼10% of cases, while Epn3 protein is overexpressed in ∼50% of BCs and correlates with poor prognosis and distant metastasis. To uncover the mechanisms through which Epn3 exerts oncogenic properties in BC we overexpressed Epn3 in non-tumorigenic mammary epithelial cells, MCF10A. We show that Epn3 overexpression causes a TGFβ-dependent EMT and expansion of the SC-like compartment, two events that are causally linked. Mechanistically, Epn3 overexpression enhances E-Cadherin endocytosis, resulting in destabilization of cell-cell junctions and translocation of β-catenin in the nucleus. The β-catenin/TCF-LEF pathway leads to the activation of the EMT transcriptional program and of the transcription of TGFβ receptors and ligands. Secretion of TGFβ ligands is a key event in the establishment of an autocrine TGFβ-positive-feedback loop, which is able to self-sustain EMT. BC cells MCF10DCIS.com and HCC1569 partially recapitulate Epn3-dependent phenotypes observed in MCF10A-Epn3 cells and Epn3 overexpression in HCC1569 cells was able to increase their in vivo tumorigenic potential upon orthotopic transplantation in NSG mice. In conclusion, our results identify new oncogenic properties for Epn3 and suggest an endocytic-based mechanism of how it contributes to EMT and BC.

Giuseppe Schiavone


Democratizing Bioethics. Online Participation and the Life Sciences
Bioethics has historically taken up the challenge of creating an arena for the adjudication of permissibility claims for practices in the broad field of the Life Sciences. Long-standing academic arguments have thus managed to percolate into proper political debates and actual policy-making. With the pressing urge to democratize politics overall, the ways in which bioethical issues have been and still are officially discussed have been thoroughly contested. A number of solutions to the alleged lack of transparency, inclusiveness and accountability in bioethical decision-making have been suggested. Some of these solutions resorted to ICTs for their implementation. However it is unclear, so far, exactly to what extent these initiatives have been able to recruit proper participation, foster reasoned deliberation, and, most importantly, cast politically legitimate decisions. Democratizing Bioethics tackles the unresolved issues of political legitimacy that underlie the current approach to deliberative public engagement initiatives for science policy-making. In doing so, it provides a political framework in which to test political theories supposed to apply to the political management of moral disagreement. Furthermore, it articulates and defends an actual political theory—moderate epistocracy for online deliberation—as a proper political means to deal with disagreement that is essentially moral arising from scientific and technical progress. Finally, the theory is preliminarily empirically tested via a tool for online direct competent participation.

Bettina Schmietow


Property redux. Ownership of human tissue and the Governance of post-genomic Research biobanks
Biobank-based genomic research is considered instrumental to realize Personalised Medicine and considerable benefits for diagnosis and therapy of many common complex diseases. It raises manifold ethical and regulatory challenges, which are discussed extensively, the main focus being on adaptations of Informed Consent to the emerging research context. This study highlights challenges to the strongly individualistic focus of classical research ethics in confrontation with biobank development. The debate on adequate protections for individual donor-participants tends towards deflationary accounts of participant rights, in which in particular the dimension of potential property in human tissue and genomic information is undervalued. Criticizing the common bioethical and legal stance that there can and should be no property in the human body and its parts, the close conceptual connections between privacy, property and consent underlying the protection of a more substantive version of participant integrity are emphasized. While ultimately the framework of traditional, in particular individual property rights is ill-suited to safeguard participant and public interests in research, property discourse is fundamental to advance discussion on the direction biobank ethics and governance should take by 1) taking serious the reordering of individual, group and societal interests in biobank research, 2) clarifying strength and limits of claims to "autonomy" and 3) refocusing to public or common goods biobank research should provide. These foundational insights are applied to an emerging normative model beneath biobank governance in which implications for the future role of consent and participant involvement in large-scale digitalized research projects are outlined.

Matteo Setti


Shedding light on Glioblastoma and derived extracellular vesicles in one clic
The treatment of glioblastoma (GBM) still represents a tremendous clinical challenge, with the average survival that is not exceeding 14 months. Given the lack of reliable prognostic markers and druggable targets in GBM, several years ago our lab’s interest focused on Chloride intracellular channel-1 (CLIC1), a protein belonging to a class of chloride channels that does not fit the classical paradigm of ion channels proteins. CLIC1 proteins can exist as both soluble globular protein and integral membrane protein with ion channel function. Upon oxidative stress, CLIC1 translocates from the cytoplasm to the plasma membrane where it exerts its function as a chloride (Cl-) channel. CLIC1 is overexpressed in several human solid tumors, including gliomas. In this study we demonstrated that CLIC1 silencing in cancer stem cells (CSCs) isolated from human GBM patients negatively influences both proliferative capacity and self-renewal properties in vitro and impairs the in vivo tumorigenic potential. Moreover, CLIC1 expression inversely associates with GBM patient survival, thus suggesting a potential exploitation of CLIC1 as a new molecular therapeutic target and a possible outcome predictor. CLIC1 has been identified as a secreted protein and detected in exosomes released from different cell types, including primary tumors. Extracellular vesicles (40-1000 nm) (EVs) are secreted by virtually all cell types that arise from the invagination and the budding of the limiting membrane of late endosomes (hence called multivesicular bodies, MVB). We showed that CLIC1 is a protein localized within EVs isolated by GBM cell lines and GBM-derived CSCs and by tuning CLIC1 expression within EVs it is possible to modulate cellular response to EVs both in vitro and in vivo. Taken together, our data suggest that CLIC1 plays an important role in regulating GBM proliferation and tumorigenic status, experimental evidences hint the possible transmission of these features to recipient cells by EV secretion.

Jacopo Sgualdino


Role of the Histone Demethylase Jmjd3 in cortical development and neural differentiation
Jmjd3 is a H3K27 demethylase that is required for the neural commitment of ESCs and controls the expression of key drivers and markers of neurulation through the demethylation of H3K27me3. Previous work from our lab has demonstrated that loss of Jmjd3 in mouse embryos causes a complex neurodevelopmental phenotype that results in perinatal death, due to respiratory failure resulting from the lack of a small network of neurons that is responsible for respiratory rhythm generation. Rescue experiments performed with catalytically active or inactive Jmjd3 have demonstrated that its demethylase activity is necessary for developing and maintaining the embryonic respiratory neuronal network. To investigate the effect of Jmjd3 loss on neocortical development, we analyzed Jmjd3 KO embryonic brains at different developmental stages using markers for stem cells and differentiated neurons. We found that loss of Jmjd3 causes an increase in the number of ventricular zone (VZ) neural precursor cells (NPCs) and a reduction in the cortical neuronal production, and that Jmjd3 KO VZ NPCs display a higher rate of cell cycle re-­‐entry together with a longer cell cycle. A detailed molecular characterization at the transcriptional level by RNA sequencing revealed that the phenotype that we observed both in vivo and in vitro in Jmjd3 KO NPCs is linked to supraphysiological activation of the Wnt/B-­‐ catenin and Notch pathways, two known regulators of the choice between self-­‐ renewal and differentiation in the VZ NPCs of the developing brains, whose upregulation has been shown to correlate with an increased proliferative potential and hampered neuronal differentiation.

Vimal Sharma


Combinatorial development of nanostructured material libraries for the study of cell substrate interaction
An in vivo cellular microenvironments in which cells are immobilized contribute an essential role in diverse cellular behavior, consisting of cellular morphology, dynamics and eventually cellular fate. Micro and nanofabrication tools widely used to mimic the in vivo surrounding to manipulate cellular microenvironment for elucidating underlying mechanisms of cellular processes. Following this concept, nanostructured (ns) material gradients and material libraries were developed in order to gain an insight of cell behavior on engineered substrates comprising Nano topography. This thesis highlights the development of gradients of nanostructured titania thin films attributing different roughness without influencing the chemical nature in order to understand the physical cues regulating primary cellular activities including proliferation and differentiation. The surface topographies were characterized by atomic force microscopy (AFM), chemical composition by energy dispersive X-ray spectroscopy (EDS) and wettability was investigated by contact angle measurements. PC12 cells derived from a Pheochromocytoma of the rat adrenal medulla were cultured on five different morphologies in presence of nerve growth factor (NGF) and the results suggest that the increasing roughness not only diminishes cellular attachment but also reduces the probability of differentiation and formation of focal adhesions. The second phase of the work targets the development of nanostructured material libraries for studying PC12 cell adhesion and growth. Here we described a combinatorial approach to construct libraries of metals comprising 54 physicochemical combinations induced by surface chemistry and topography. In addition the surface properties including surface topography, surface chemistry, and wettability were characterized followed by investigating cellular behavior influenced by different physicochemical conditions of nanostructured material libraries.

Lara Sicouri


Tumor suppressor mutagenesis driven by DNA deaminase
Genomic instability is commonly associated with pathological disorders including cancer. The progressive accumulation of genetic abnormalities in cancer-associated genes can confer cellular autonomous proliferation, contributing to the oncogenic transformation. Although animal models have been instrumental to the understanding of the molecular mechanisms responsible for tumor progression, there are severe limitations for success. This thesis work aimed at the development of an innovative animal model that could recapitulate the ongoing lifelong accumulation of DNA lesions, leading to neoplastic transformation. To this end, the natural DNA mutating enzyme AID was fused to sequence specific DNA binding proteins. They were engineered to target tumor suppressor genes and to induce low-frequency mutagenesis in cell lines and in zebrafish. A TALE-Aid fusion protein was targeted to the p53 locus of mouse cell lines, and the Aid-dependent mutations were monitored by next-generation sequencing. The induced mutations occurred at a comparable frequency to those observed in Aid-induced non-immunoglobulin gene targets in B cells. Mutations were found mostly in the DNA binding domain of p53, possibly reflecting AID- hotspot residues in p53. Our approach also induced mutations that have not been characterized previously, and could provide further insight into p53 dependent oncogenesis. When TALE-AID were expressed and targeted to the p53 locus of zebrafish embryos, the activity induced developmental abnormalities with variable severity, leading to both an increased mortality, and impaired ovarian maturation and fertility. We have developed and initially characterized a novel tool for the in vitro / in vivo study of the accumulation of mutations in cancer-associated genes.

Gianluca Sigismondo


ChroP approach dissects the dynamic profiling of chromatome at enhancers of inflammatory genes
Inflammatory stimuli drive a fine rearrangement of cell-specific chromatin determinants at cis-regulatory regions of inflammatory genes. Although a few determinants are known (i.e. H3K4me1 and Pu.1), a global picture of the enhancers’ molecular signature and how specific factors dynamically synergize during the inflammatory response remain incomplete. The aim of my thesis is the global characterization of the enhancers’ determinants and their dynamic profiling during inflammatory response. I addressed this issue by employing the chromatin proteomics approach (ChroP), which combines chromatin immunoprecipitation and mass spectrometry-based proteomics, to dissect histone post-translational modifications (modificome) and chromatin-binding proteins (interactome) associated with a specific chromatin region. We used H3K4me1 and Pu.1 antibodies to specifically enrich enhancers from macrophage-derived RAW264.7 cells. Native chromatin and formaldehyde-fixed chromatin from SILAC labeled cells were used for the hPTMs profiling and the identification of chromatin binding proteins, respectively. Furthermore, a triple-SILAC setup was used in time-course experiments to profile the gene transcriptional activation triggered by lipopolysaccharide. Our findings suggest that enhancers have in basal condition an overall higher-ordered structure that is maintained during the inflammation; while a subset of proteins displays a dynamic behavior, i.e. the PBAF complex and Dnmt1, that seem to synergize with specific hPTMs to set an environment permissive to transcription. Interestingly, a number of factors are newly recruited at enhancers, suggesting a role in fine-tuning the appropriate gene expression: among them, Junb and Stat1 are well characterized to be involved in inflammatory response, while others (i.e. DDIT3 and Ifi204) are novel, very promising and under further investigation.

Shruti Sinha


Detection of structural variations during liver cancer progression
Hepatocellular carcinoma (HCC) is one of the most frequent and lethal cancers and accounts for 70-80% of all liver cancer. HCC is almost invariably associated with an underlying inflammatory state, whose direct contribution to the acquisition of critical genomic changes is unclear. We mapped the acquired genomic alterations in the exome of human and mouse HCCs induced by defects in hepatocyte biliary transporters, which cause the onset of chronic inflammation that develops into cancer even in the absence of external causative agents. In both human and mouse cancer genomes we found few somatic point mutations with no impairment of cancer genes, which is very different from other HCCs studied so far. To identify copy number variations directly from exome sequencing data, we developed a novel method that combines the frequency of heterozygous germline mutations with the read depth to identify genes undergoing deletion, amplification and LOH events. We applied this method to mouse exome re-sequencing data and observed the acquisition of massive gene amplification and rearrangements in both species. Copy number gains preferentially occurred at late stages of cancer development and frequently targeted the MAPK signaling pathway, and in particular direct regulators of JNK. The pharmacological inhibition of JNK retarded cancer progression in the mouse. Our study demonstrates that HCC induced by bile acids and inflammation develops through genomic modifications that can be clearly distinguished from those determined by other etiological factors, such as virus and alcohol.

Elisa Sogne


Novel Platform for Biosensing Application Based on Cluster-Assembled Materials
Rapid methods to identify bacteria in biological samples are important for prompt antimicrobial therapy. The current detection methods are classical biological sample cultures and biochemical tests, which are however, time-consuming and not highly sensitive. A novel and highly performing approach is offered by aptamers acting as recognition elements able to detect epitopes on the surface of a bacterium. Aptamers interacting with specific bacteria are known and then could provide a solid base for developing promising solutions for this issue. With this PhD work I intended to tackle one drawback of aptamer-based biosensor: the lack of platforms for high density aptamers immobilization. Cluster-assembled thin films, have been optimized as supports to demonstrate that aptamers, targeting Staphylococcus aureus, well adhere on these substrates and keep their functionality. Moreover, the characteristics of the nanostructured zirconium oxide thin film: thermal stability, good reactivity towards -OH and -COOH groups and nano-morphology, make this material a suitable candidate for the realization of platforms for general screening and biosensing applications. This strategy will offer a promising way for the development of an user-friendly aptamer-based biosensors for screening biological samples. Furthermore, I focused on a technological problem, regarding the need of substrates to perform correlative light-electron microscopy (CLEM), designing, developing and testing a device which improve the feasibility of correlative fluorescence/confocal and scanning electron microscopy.

Ilaria Spadoni


Identification and characterization of the "gut vascular barrier"
In order to protect the body from harmful environmental agents, the intestine has developed a physical barrier formed by the epithelium and a specialized immune system able to induce either tolerance against food antigens and intestinal flora or inflammatory responses against dangerous microorganisms. Here, we describe a new barrier that we called the gut vascular barrier (GVB) that plays a fundamental role in controlling the spreading of molecules and bacteria to systemic sites. We found that intestinal endothelial cells (ECs), located beneath the epithelium, expressed the main components of tight and adherent junctions necessary to avoid paracellular trafficking of molecules. In addition, we observed the existence of a "gut vascular unit" whereby ECs were associated with enteric glial cells and pericytes, whose role in the establishment of the GVB remains to be analyzed. We showed that GVB integrity could be modified by Salmonella typhimurium infection. Indeed, upon infection ECs up-regulated the expression of PLVAP, marker of damaged/immature vascular barrier, which correlated with a higher permeability of the endothelium. One way by which S. typhimurium could modify the barrier properties of the intestinal ECs was through the negative regulation of the Wnt/β-catenin signaling pathway. Indeed, we found that β-catenin activation was reduced upon infection in vitro. Consistently, Salmonella was incapable to modify ECs permeability and to spread systemically in mice where β-catenin was constitutively activated by genetic means only in vascular ECs. Furthermore, we demonstrated that Salmonella pathogenicity island-2 was involved in the regulation of Wnt/β-catenin signaling pathway in ECs.

Valeria Spadotto


The role of arginine methylation in miRNA biogenesis investigated by MS-proteomics
MicroRNAs (miRNAs) are short, non-coding RNA molecules that fine tune gene expression at the post-transcriptional level. The efficient miRNA processing is fundamental to maintain their correct levels in the cell. Alterations in microRNA biogenesis occur in different pathological conditions, including cancer. Hence, dissecting the underlying molecular mechanisms is crucial to understand how aberrant miRNA patterns are generated. The processing step operated by the Large Drosha Complex (LDC), fundamental and rate limiting for the production of miRNAs, occurs in the nucleus and consists in the cleavage of the primary (pri-) into the precursor (pre-) miRNAs. In a previous MS-based analysis of the cellular methylome, we showed that protein methylation occurs on the majority of the LDC subunits, suggesting the possibility that this post-translational modification could contribute to the regulation of miRNA biogenesis. To investigate the role of protein methylation in the regulation of the LDC, we characterized the methylated sites on the subunits of the complex, thus generating the first comprehensive and high-confident methyl-proteome of the LDC. By modulating the expression of PRMT1 we analyzed how methylations change at specific sites of distinct LDC subunits, thus identifying PRMT1 substrates. The alteration of the LDC-methylation state upon PRMT1 depletion correlated with an impairment of the pri- to- pre-miRNA processing, which resulted in the massive deregulation of mature miRNA expression. Although the molecular mechanism remains to be elucidated, the results described in this thesis uncover a key role of arginine methylation in the regulation of the LDC activity and, consequently, in miRNA biogenesis.

Vanessa Spagnolo


Effect of fasting mimicking conditions on hormone therapy efficacy and resistance acquisition in breast cancer
The most common cancer diagnosed among women worldwide is breast cancer (BC) with 60-80% belonging to the ER+BC subtype. Development of resistance is the major driver of ER+BC-related death. Well-tolerated treatments that allow concomitant targeting of the heterogeneous cancer cells are needed in order to abolish resistance. Fasting and fasting mimicking diets (FMD’s), acting in a broad way, have shown great potential to protect the healthy cells of the body during chemotherapeutic treatment, while augmenting treatment efficacy, mainly through lowered levels of blood glucose and circulating Insulin-like growth factor-I (IGF-I) availability. In ER+BC the interconnection of the ER signalling and the IGF1R signalling builds up a strong network difficult to target with specifically acting drugs. This work provides evidence that cycles of a vast acting FMD have the potential to postpone development of resistance to specific acting endocrine therapy (ET) and circumvent development of resistance to the combined treatment of ET and non ER+BC-specific drugs in vivo. FMD’s capacity to abolish resistance exists if applied at the onset and but also at later stages when insensitivity to these drugs has been acquired. Concomitant and early targeting exhibits a killing effect in vitro and in vivo, potentially on resistance driving subpopulations.

Aishwarya Subramanian


Mitochondrial MARCH5 ubiquitin ligase abrogates MCL1-dependent resistance to BH3 mimetics via activation of NOXA
BH3 mimetic compounds induce tumor cell death through targeted inhibition of anti-apoptotic BCL2 proteins. Resistance to one such compound, ABT-737, is due to increased levels of anti-apoptotic MCL1. Using chemical and genetic approaches, we show that resistance to ABT-737 is abrogated by inhibition of the mitochondrial RING E3 ligase, MARCH5. Mechanistically, this is due to increased expression of pro-apoptotic BCL2 family member, NOXA, and is associated with MARCH5 regulation of MCL1 ubiquitination and stability in a NOXA-dependent manner. MARCH5 expression contributed to an 8-gene signature that correlates with sensitivity to the preclinical BH3 mimetic, navitoclax. Furthermore, we observed a synthetic lethal interaction between MCL1 and MARCH5 in MCL1-dependent breast cancer cells. Our data uncover a novel level at which the BCL2 family is regulated; furthermore, they suggest targeting MARCH5-dependent signaling will be an effective strategy for treatment of BH3 mimetic-resistant tumors.

Olga Tanaskovic


Lack of p21 expression in tumor-associated APCs triggers the activation of a potent anti-tumor immune response
Over the last decade, the cell-cycle inhibitor p21 has been shown to sustain leukaemia propagation with two distinct mechanisms. On one hand, p21 was shown to be critical for maintaining increased self-renewal capacities of leukaemia stem cells. Indeed, the absence of p21 in leukaemia stem cells leads to their functional exhaustion, which results in loss of leukaemia transplantability in syngeneic mice. On the other hand, p21 expression is crucial for evading the surveillance mechanisms of the immune system, thus ensuring tumor growth. Specifically, lack of p21 in the leukemic microenvironment activates a potent CD4+ T-cell mediated immunological response against tumor in syngeneic context (unpublished data from the host laboratory). To translate the observed p21- dependent anti-tumoral immunity into novel immune-therapies against cancer, underlying mechanisms needed to be unrevealed. In my thesis work, I dissected the cellular bases of the p21-dependent anti-tumor immunity. I disclosed a crucial role of the p21-/- tumor microenvironment in triggering activation of an anti-tumor immunological response. In particular, for the first time I identified rare iron-loaded CD68+ tumor- associated macrophages (iTAMs) in the p21-null context as key mediators of a potent immunological mechanisms of cancer clearance. By unravelling crucial players of the p21-dependent anti-tumor immunity, my work set the basis for the future design of novel anti-cancer vaccines. Such vaccines will grant more efficient and less toxic treatment for cancer patients.  

Alessandra Tesi


Gene regulation by Myc during B cell activation
c-­‐Myc is a transcriptional regulator required for the cellular response to proliferative stimuli. The gene expression programs regulated by Myc in physiological settings remain to be clarified. Here, we provide a complete characterization of Myc-­‐dependent regulatory events in primary mouse B cells following activation by bacterial lipopolysaccharide (LPS). Taking advantage of cells homozygous for a conditional knockout allele of c-­myc, we induced deletion before LPS stimulation, followed by genome wide profiling of mRNA levels and Myc-­‐DNA interactions. In contrast with previous studies, in which Myc was proposed to directly drive transcriptional amplification at all active loci (Nie et al. 2012, Lin et al. 2012), our study revealed that Myc is required for the up-­‐and down-­‐regulation of distinct subsets of genes. Early after stimulation, occurring prior to the global increase in RNA production. These gene expression programs where partially overlapping with those regulated by Myc upon oncogenic activation, a distinction made not only in B‐cells, but also in fibroblasts (Sabò et al., 2014, Perna et al. 2012). Our data also show that Myc dependent regulation can occur at the level of RNA Polymerase II loading, as well as elongation. Altogether these data provide an extensive picture of Myc' saction in response to a mitogenic stimulus, highlighting the importance of Myc‐target genes in the remodeling ofcellular physiology and metabolism.  Systematic work will be needed to unravel which, among allthe Myc-regulated genes, are critical in mediating this chain of events.    

Alessandro Tocchio


Design of vascularizable scaffolds for large tissue engineering
The emerging field of tissue engineering is dedicated to restore, maintain or improve the functions of damaged or lost human tissues. However, despite significant successes have been achieved over the last 20 years, several challenges still remain, preventing a pervasive clinical application of tissue engineering. One of the main challenges lies in the development of scaffolding materials able to mimic the complex organization of the in vivo milieu and provide tailored stimuli for tissue growth and maturation. A fundamental aspect of this problem resides in the design of scaffolds having three-dimensional vascular architecture, able to provide optimal nutrients diffusion, supporting and maintaining viable tissue in vitro, and capable to promote vascularization after implantation. The lack of proper vascularization is currently limiting the size of the engineered tissues to smaller than clinically relevant dimensions. The aim of this PhD work, merging the study of novel biomaterials and the development of original microfabrication methods, is to create enabling technologies towards the design of innovative scaffolds for large tissues engineering. For this purpose, a library of RGD-mimetic hydrogels with controlled chemical, mechanical and biological features has been developed. The obtained hydrogels have been combined with foaming and sacrificial molding techniques to engineer customizable scaffolds with hierarchical three-dimensional architectures. These novel hydrogel scaffolds supported optimal three-dimensional cell growth and promoted in vitro vascularization in large constructs. The reported results suggest that the presented approach could represent a viable solution to scale engineered tissue to clinically relevant dimensions releasing the full potential of regenerative medicine.

Emiliana Tognon


The study of the role of ESCRT-0 in Notch signaling and tumor suppression
Sorting and degradation of ubiquitylated cargoes depends on the endosomal sorting required for transport (ESCRT) machinery. The ESCRT machinery is composed of four multi-subunit ESCRT complexes (ESCRT-0, -I, -II, -III), which act in a sequential fashion to deliver endocytic cargoes into the internal luminal vesicles (ILVs) of the multivescicular endosome (MVE) for subsequent degradation. ESCRTs sort a number of transmembrane proteins including Notch and the JAK/STAT signaling receptor Domeless. In Drosophila epithelial tissue, mutation in ESCRT –I, -II, -III components results in misregulation of several signaling pathways, loss of epithelial polarity and unrestrained proliferation, suggesting that ESCRT genes act as tumor suppressors. Unexpectedly, Drosophila Hrs, one of the two components of the ESCRT-0 complex that acts upstream of the other ESCRT complexes have been found to be dispensable for tumor suppression. Thus, when I started my Ph.D. it was unclear whether ESCRT-0 had a tumor suppressive function. In my first project, I have found that mutation of Stam, a second ESCRT-0 component or of both Hrs and Stam result in accumulation of ubiquitinated proteins and of the signaling receptors Notch and Domeless. Nevertheless, mutant tissue displays normal tissue architecture, proliferation and Notch signaling activation. Overall, our in vivo data indicate that the ESCRT-0 complex does not play a crucial role in tumor suppression. 1.1         Abstract Project 2 In mammals, the Transcription Factor EB (TFEB) family of basic Helix-Loop-Helix (bHLH) transcription factors regulates both lysosomal function and organ development. However, it is not clear whether and how these two processes are interconnected. In Drosophila, the Microphthalmia-associated Transcription Factor (Mitf) is the unique homolog of the TFEB family. In my second project I have found that Mitf acts similar to its mammalian counterparts as transcription factor shuttling from lysosomes to the nucleus to regulate V-ATPase expression and lysosomal biogenesis. Interestingly, I found that V-ATPase subunits display diverse expression patterns in the wing imaginal disc, suggesting complex regulation of V-ATPase during development. Remarkably, I could show that Mitf cooperates to regulate expression of a key component of the V-ATPase during differentiation of proneural clusters (PNCs), a process that specifies cells with neuronal identity. In addition, I have observed that the PNCs possess a distinctive endo-lysosomal compartment and Notch localization. Finally, I have determined that modulation of V-ATPase and Mitf in the disc alters endo-lysosomal function and PNC development. Overall my in vivo analysis indicates that lysosomal-associated functions regulated by V-ATPase/Mitf axis might play a role in tissue patterning during Drosophila development.

Claudia Tonelli


Genome-wide analysis of p53-dependent programs in tumor suppression
The transcriptional programs triggered by p53 during tumor suppression and in response to DNA damage remain to be clarified. Using whole genome mapping of p53 binding and gene expression profiling, we investigated the transcriptional circuitry induced by p53 in suppressing cancer development and in response to genotoxic injury. We studied the progression of Myc-induced lymphomas in Eμ-myc transgenic mice, as well as the regression of these lymphomas following restoration of p53 function, by either pharmacological or genetic means. In parallel, we determined the p53-dependent transcriptional program in splenic cells from mice exposed to ionizing radiation. We thus expanded our understanding of the p53 response to oncogenic and genotoxic stress and identified a set of novel components of the p53 transcriptional program. Currently, we are testing the impact of these new p53 target genes on tumorigenesis using an RNA interference (RNAi)-based functional genetic screen. Altogether our data represent an extensive characterization of the p53-regulated network in response to different stimuli and will hopefully highlight new tumor suppressive mechanisms, paving the way for their therapeutic application.

Chiara Tordonato


A novel microRNA family as molecular determinant in mammary stem cells
MicroRNAs (miRNAs) are an evolutionarily conserved class of small (18-22 nucleotides) noncoding RNAs involved in the regulation of a variety of cellular and developmental processes. MiRNAs recently emerged as key regulators of stem cells (SCs) cell-fate. In breast, forced expression of miR-200 or let-7 could inhibit both normal and cancer SCs by silencing self-renewal determinants. Intriguingly, these miRNAs are poorly, or not, expressed in the SC compartment and their function is mainly achieved through the induction of differentiation, suggesting that other miRNAs could be required specifically in SCs to maintain their identity. We aimed at identifying those miRNAs specifically expressed in mammary SCs by exploiting an innovative technique for the isolation of quasi-pure SCs/CSCs. We, thus, identified a miRNA family highly enriched in SCs isolated from both normal and cancer human primary samples and mouse mammary epithelia. High levels of this miR-family were also found in CSCs obtained from human primary biopsies and correlated with tumors of the basal subtype and with an elevated CSC content. Depletion of these miRNAs impaired self-renewal of both normal SCs/ and CSCs, as measured by serial mammosphere assay and by in vivo tumorigenicity assay. We employed a global transcriptomic approach to assess the effects of the loss of this miRNAs in stem cell biology, coupled with target predictions. We revealed that the gene expression program of “stemness” is specifically dependent on the expression of this miR-family and regulated by the interaction of these miRNAs with multiple genes critical for self-renewal and cell fate determination.

Eleonora Valentini


Understanding the catalytic mechanisms of ubiquitin-E3 ligases
E3 ubiquitin ligases are regulatory enzymes of the ubiquitination pathway, they are responsible for substrate specificity. This thesis aimed at deciphering the molecular mechanisms through which two different E3 ligases, Nedd4 and Rabex-5, exert their activity. Nedd4 is the prototype for HECT-E3 ligase while Rabex-5, containing an A20 zinc finger domain (ZnF_A20) instead of a canonical RING, could be defined as an atypical RING-E3 ligase. For Nedd4, we provided the first crystal structure of the catalytic intermediate of HECTNedd4~Ub in complex with Ub non-covalently bound to the UBD present in the N-lobe of HECTNedd4. Our structure represents the next step of the transfer of UbD from catalytic cysteine of E2 to the one of E3 in which the UbD C-terminal tail is in an extended conformation primed for catalysis. Our data strongly supports the sequential addition model proposed for HECT proteins. We clarified some aspects of Rabex-5 as E3 ligase. By Y2H, GST-pull-down assays and ITC analysis, we identified specific E2 partners, Ube2D and Ube2E families, that bind Rabex- 5 only when in active Ub-loaded state. Performing IVT-auto-Ub assay and disulfide stability assay we confirmed that ZnF_A20 is the minimal domain responsible for catalytic activity. To obtain the structure of Rabex-51-74:E2-Ub complex, we tested, unfortunately without success, crystallization trials and SAXS analysis. We also analyzed Rabex-5 catalytic activity towards on H-Ras, unique substrate of Rabex-5 so far identified, and we disproved that H-Ras is a Rabex-5 substrate. To identify candidate substrates we profiled 20.000 human proteins using a microarray-based ubiquitination screening. List of 67 proteins represent the most statistically stringent and conservative estimate for Rabex-5 substrates that we will validate.

Shikha Vashisht


Computational Approaches in the Estimation and Analysis of Transcripts Differential Expression and Splicing: Application to Spinal Muscular Atrophy
Spinal Muscular Atrophy (SMA) is among the most common genetic neurological diseases causing infant mortality. SMA is caused by deletion or mutations in survival motor neuron 1 gene (SMN1), which are expected to generate mRNA-splicing alterations and reductions in mRNA-transport within the motor neurons (MNs). SMA ultimately results in the selective degeneration of MNs, but the underlying reason is still ambiguous. The aim of this study is to investigate splicing abnormalities and to identify genes presenting differential splicing events, possibly involved in SMA pathogenesis at genome-wide level. We performed RNA-Sequencing data analysis on two SMA-patients and two healthy-controls, with 2 biological replicates each sample, derived from their iPSC-differentiated-MNs. Three analyses were executed. Firstly, differential expression analysis was performed to identify possibly mis-regulated genes. Secondly, alternative splicing analysis was conducted to find differentially-used exons (DUEs), as splicing patterns are notably altered in MNs by SMN protein suboptimal levels. Thirdly, we did RNAbinding protein (RBP) - motif discovery for an identified set of DUEs, to pinpoint MN-specific mechanisms underlying such alterations. Gene Ontology enrichment analysis of identified significantly differentially expressed genes and DUEs revealed various GO-terms related to axon-guidance, musclecontraction, microtubule-based transport, axon-cargo transport etc. suggesting their involvement in SMA. We obtained promising results from motif analysis that has identified 22 RBPs where 5 RBPs from the PABP family are known to interact with SMN protein, enhancing mRNA-stabilization and mRNA-transport in MNs. To validate our results wet-lab experiments involving precise recognition of RNA-binding sites are required which might offer potential therapeutic role towards treating SMA. Further we observed, the current methods for an effective understanding of differential splicing events within the transcriptomic landscape are insufficient. To address this problem, we developed a computational model having a potential to precisely estimate “transcript expression levels” within a given gene locus by disentangling mature and nascent transcription contributions for each transcript at per base resolution. Exonic and intronic read coverages were modeled and transcript expressions were estimated, which best approximated the observed expression in total RNA-Seq data. Our model has an application in the detection of differential splicing events. At exon level, differences in the ratio of the sum of mature and the sum of nascent transcripts over all the transcripts in a gene locus gives an indication of differential splicing. We have implemented our model in R-statistical language.

Pietro Vella


Transcription factors, histone and DNA modifications: old and new actors on the chromatin stage
During my Ph.D. I have carried out two projects. The first one was focused on the Yin Yang1 (YY1) transcription factor, an essential regulator of mouse embryogenesis, with the aim to characterize its role in Polycomb recruitment to chromatin in mammalian cells. In this project I showed that in mouse embryonic stem cells Yy1 has genomewide Polycomb independent activities and binds chromatin in close proximity of the transcription start site of highly expressed genes. In general Yy1 has a dual transcriptional regulatory function, whereas it positively influences biogenesis of several small noncoding RNAs. Finally, Yy1 extensively shares genome wide binding properties with Mycrelated transcription factors and their coordinated binding at promoters potentiates gene expression, proposing YY1 as an active component of the Myc transcription network that links embryonic stem cells to cancer cells. The second project was aimed to elucidate the nuclear functions of the Olinked N acetylglucosamine (O-GlcNAc) transferase (Ogt), an enzyme essential for mouse development that catalyzes glycosylation both on nuclear and cytoplasmic proteins in several cell types. My data found that the teneleven translocation (TET) proteins Tet1 and Tet2 are the main partners of nuclear Ogt in ES cells. At a genomewide level, Ogt preferentially associates with Tet1 in close proximity of transcription start sites rich in cytosine/guanine dinucleotides (CpG). These regions are characterized by low levels of DNA modification, suggesting a link between Tet1 and Ogt activities in regulating CpG island methylation. Finally, Tet1 functions as the main chromatin recruiter of Ogt, which in turns regulates Tet1 activity. Taken together, these data characterized how Olinked glycosylation is recruited to chromatin and interacts with the activity of 5methylcytosine hydroxylases.

Claudio Vernieri


Cdk1 kinase counteracts PP2ACdc55 phosphatase to induce APC/C phosphorylation and adaptation to the Spindle Assembly Checkpoint (SAC) in budding yeast
The spindle assembly checkpoint (SAC) monitors that all sister chromatids are correctly attached to microtubules of the mitotic spindle during prometaphase. The correct attachment is known as biorientation, and it is the prerequisite for proper partitioning of the duplicated DNA from the mother to the two daughter cells. Until the last kinetochores are bioriented, the SAC arrests progression into anaphase by inhibiting the Anaphase Promoting Complex or Cyclosome (APC/C) bound to its coactivator Cdc20, therefore stabilizing the cohesin rings that hold pairs of sister chromatids together. When the SAC is continuously activated, cells remain arrested in prometaphase for some hours, but not indefinitely; even if the checkpoint is not satisfied, eventually cells separate the duplicated DNA material and progress into anaphase. This phenomenon is known as adaptation to the SAC and is poorly understood from the molecular viewpoint. By using budding yeast (S. cerevisiae) as a model organism, in this work I show that adaptation to the SAC requires phosphorylation of the APC/C, which is stimulated by the Cyclin dependent kinase 1 (Cdk1) bound to its mitotic regulatory subunit, Clb2, and is opposed by the phosphatase PP2ACdc55. I propose that PP2ACdc55 and the APC/C are implicated in a double negative feedback loop of reciprocal inhibition, which regulates transition into anaphase in adapting cells: when accumulating Clb2 provides sufficient Cdk1:Clb2 activity to allow initial activation of APC/CCdc20, the antagonist of the APC/C, PP2ACdc55, starts to be inhibited. This strongly reinforces APC/CCdc20, and leads to a rapid and irreversible transition into anaphase.

Chiara Villa


Microencapsulation of pancreatic islets for cell transplantation in type 1 diabetes without immunosuppression
ype 1 Diabetes mellitus (T1D) is a chronic autoimmune disease caused by the attack of T lymphocytes on pancreatic β-cells, leading to absolute insulin deficiency. For T1D patients, exogenous insulin injections are a lifesaving treatment, but it does not prevent hypoglycemia daily risk. Pancreatic islet transplantation is one of the therapeutic approaches in case of severe hypoglycemic unawareness, but it requires life-long systemic immunosuppression, causing several side effects. Moreover, most patients show insulin-independence and early graft function lost, due to immunological and physiological limitations of the transplant site. Alginate encapsulation represents an alternative providing a physical barrier from the host immune system attack, and it has been evaluated in pre-clinical and clinical settings. The successful use of alginate microcapsules has been hampered by 1) the large diameters, and 2) the mechanical instability. The large capsule size constitutes an oxygen and nutrient diffusion barrier and limits the choice of transplant sites to areas not conceived for cell survival, such as the peritoneal cavity. Conventional alginate microencapsulation has been optimized by minimizing capsule size (450-550μm in diameter), increasing the cell loading density (nearly 3%), and by using highly biocompatible Ultra-Pure medium viscosity sodium alginate (UP-MVG). This allowed for transplantation of microencapsulated islets in the well-vascularized epididymal fat pad (EFP), engineered with a fibrin matrix to promote graft engraftment. Under physiological conditions alginate capsules show swelling and pore size increase. To protect the cells, the capsule must therefore be carefully designed, especially with respect to mechanical stability. Thus, novel alginate-based capsules have been designed to improve in vivo stability of alginate: 1) hybrid microcapsules (MicroMix) using an electrostatic droplet generator method by mixing UP-MVG with Polyethylene Glycol (Peg) 2) UP-MVG microcapsules double coated (Double) with Peg through an emulsification process. One of the main challenges has been to ensure in vivo long term alginate capsule stability, with the hope that a complete knowledge of alginate-based capsule biocompatibility, mechanical properties and permselectivity will be useful for successful clinical transplantation.

Alessandro Vitriolo


Multi-omic deconvolution of the regulatory networks underlying neurodevelopmental and autism spectrum disorders: a multidimensional analysis for a new disease modelling paradigm
Multi-omic deconvolution of the regulatory networks underlying neurodevelopmental and autism spectrum disorders: a multidimensional analysis for a new disease modelling paradigm   Recent literature has highlighted that mutations causing neurodevelopmental syndromes are particularly enriched in genes related to chromatin. Most of such disorders couple different shades of intellectual disabilities with peculiar cranio- facial features and systemic defects which turned out to be shared, opposite or unique across them. Building on our expertise to harness potency and stability of induced pluripotent stem cells (iPSCs), we identified two main axes of development through which we characterized on the one hand cerebral cortex related dysregulations and on the other hand cranio-facial features associated traits, peripheral nervous system- and cardiovascular system-related dysregulations. First, we identified transcriptional modules and disease-specific dysregulations reverberating from the pluripotent stage to disease-relevant tissues. Then, we identified a set of putative direct targets of PRC2 complex involved in early brain development and BAZ1B-specific transcriptional dysregulations in neural crest stem cells, that confirm its importance for migration and craniofacial morphogenesis but more in general for chromatin remodelling and modern human evolution. Finally, we produced a molecular characterization of KMT2D function in adult cortical neurons. In doing so we developed new tools and standards for NGS-based differential expression analyses, and an analytical framework to boost the ability of identifying the effect of knocking down a certain gene on transcriptional and epigenetic landscapes.

Thalia Vlachou


Functional and genetic heterogeneity in acute myeloid leukaemia
Acute myeloid leukaemia (AML) is the most frequent leukaemia in adults, and still represents a disease with an unmet medical need, with 50-60% of patients relapsing within 3 years after diagnosis. AMLs are characterised by a high degree of intra-tumour heterogeneity, both at the biological and the genetic level, which is critical for tumour maintenance and response to treatments. Biologically, AMLs are organised hierarchically, with rare stem-like cells (leukaemia stem cells, LSCs) endowed with the unique properties of self-renewal and differentiation. Genetically, AMLs harbour patient-specific combinations of different driver mutations, which are organised within individual cases in sub-clones with distinct growth properties. We hypothesized that tumour maintenance and relapse in AMLs are driven by the selective expansion of quiescent sub-clones within the LSC population, which serve as the genomic and functional reservoir of the tumour. The experimental strategy we employed to test this hypothesis is based on the xenotransplantation of human leukaemias, the implementation of an in vivo clonal tracking approach, the functional isolation of leukaemic subpopulations with diverse proliferation histories and whole-exome sequencing (WES) of bulk and isolated leukaemic subpopulations. Our aims were to assess the proliferative hierarchy of LSCs and to examine their intrinsic genetic heterogeneity. We identified two functional LSC classes, quiescent and cycling, that are in equilibrium in the tumour and largely share the same clonal architecture. We further observed that genetic leukaemic clones appear to consist of a high number of individual LSCs, the majority of which exhaust upon serial transplantation. Finally, by genetic analyses of isolated leukaemic subsets, we were able to detect a specific enrichment for rare mutations in the quiescent compartment of two patient xenografts. Our data indicate that tumour evolution is sustained by the quiescent LSC pool and suggest that their highly proliferating counterpart has a finite lifespan. We expect that 14 the results of our studies will provide new insights into the mechanisms of disease progression and treatment response in AML, and potentially reveal novel therapeutic approaches.

Nadine Wossner


Characterization of the role of USP25 in EGFR endocytosis
Deregulated EGFR signalling is a significant feature in different stages of oncogenesis and it contributes to several cancer types. One important mechanism whereby cancer cells can obtain increased and uncontrolled EGFR signalling is to escape down-modulation of the receptor. Ubiquitination of EGFR and of members of the endocytic machinery has a key role in this process, regulating receptor internalization, trafficking and degradation. Deubiquitinating enzymes (DUBs) can reverse the ubiquitination process, antagonizing or even promoting receptor degradation. To identify DUBs altering EGFR degradation we undertook a genome-wide small interfering RNA screen targeting all known active DUBs. We identified twelve novel DUBs affecting EGFR degradation by using immunoblot-based approaches complemented by an ELISA-based assay. Among them USP25, a member of the ubiquitin-specific protease (USP) family, displayed one of the strongest effects. We found that the degradation rate of EGFR is enhanced upon USP25 knock-down. Quantitative internalization assays revealed that depletion of USP25 leads to a faster internalization rate of EGFR. Consistently, overexpression of wild-type USP25, but not its catalytic inactive mutant, partially blocked EGFR internalization. Furthermore, we scored an increase in the EGFR ubiquitination upon USP25 knock-down, in particular at early time points post EGF stimulation, suggesting that EGFR is a direct target of USP25. Taken together our study identifies USP25 as a novel negative regulator of EGFR ubiquitination and endocytosis, involved in early internalization events. USP25 may represent a suitable “druggable” target for pathological conditions where EGFR is deregulated and opens up a promising direction for future investigations.

Yunsong Yan


Novel Electroactive Soft Actuators Based on Ionic Gel/Gold nanocomposites produced by supersonic cluster beam
In this thesis, I demonstrated the fabrication of electroactive IGMNs based on ionic conductive PAA-co-PAN network with embedded HNC by mean of SCBI. The production of cluster-assembled gold thin films (100 nm thick), inter-penetrating with the quasi-solid state electrolyte, enabled the manufacturing of elastic soft actuators provided with flexible electrodes having low surface electric resistance and large surface area for an effective charge storage. The electro-responsive properties of the materials are intermediate between that of IPMCs and bucky-gels. This feature is due to the synergistic effect of the quaternary ammonium TEA+ counter-ion and HNC on the transport ability and electrochemical characteristics of the functional nanocomposites. Both the organic cation and the inorganic nanostructures confer high ionic conductivity (0.35 mS/cm) and large double layer capacitance (30 μF/cm2) to the composites, by favoring the partial dissociation of the EmimBF4 ionic liquid incorporated in the structure (solvent molar uptake is 37%). The materials showed sensitivity to the applied electric field starting at 0.1 V and exhibited high-performance actuation in response to electrical stimuli at low voltages (up to 1.04% net strain at 5 V) and good durability in their frequency response (up to 76000 cycles at 2 V and 1 Hz). The manufacturing process adopted for the actuators fabrication is cost-effective and suitable for mass production. The novel IGMNs presented here represent a forefront technological solution for the development of smart flexible components suitable for the manufacturing of soft bio-mimetic robots, deformable electronic and haptic systems, and wearable biomedical devices.

Martina Zobel


Spatial regulation of signalling by the endocytic protein NUMB
NUMB is a cell fate determinant that controls signaling outputs during asymmetric cell division. NUMB is a tumor suppressor protein which regulates NOTCH and p53 signalling pathways. Its loss occurs in 50% of all breast tumors and 30% of lung cancers and correlates with increase aggressiveness and poor prognosis. NUMB binds to key players of Clathrin-mediated endocytosis, regulating the internalization of various receptors and it has been found to localize in recycling endosomes where it may regulate the delivery of cargos back to the plasma membrane (PM). We found that NUMB is as a negative regulator of Circular Dorsal Ruffles (CDRs) formation downstream of c-MET and PDGFR. This is accompanied by an increased mesenchymal mode of motility and cell invasion. CDRs formation depends on the recycling back of active RAC to spatial restricted sites of PM via ARF6. We found that NUMB is enriched in ARF6 recycling compartments and inhibits MHC I and RAC recycling. These evidences suggest that NUMB might act as a negative regulator of ARF6-dependent recycling. We found that among the ARF6 GEFs, only EFA6B contains an NPxF motif in its N-terminal domain and directly binds the NUMB PTB domain. Functionally, the removal of EFA6B by itself does not alter CDR formation, but it inhibits the increase of CDR formation that is brought about by NUMB down-regulation. Our data suggest that NUMB may regulate the ARF6-dependent recycling pathway, which is required for migration and invasion of tumor cells, by interacting with and possibly modulating its GEF EFA6B.