Among females, breast cancer is the most common diagnosed cancer and the leading cause of tumour death. In the last decade, it is notable a decreasing trend thanks to presence of primary prevention programs, of diagnostic screening (mammogram) and of new therapeutic strategies. At present, the broad heterogeneity of breast cancer reflects the well-accepted notion that there is not just one disease with disparate variant subtypes, but that breast cancer instead represents a collection of distinct neoplastic diseases of the breast and the cells composing it. Behind this complexity, several systems have been developed to classify this very highly heterogeneous disease and possibly to get information about tumour behaviour and provide more effective therapies. Histological classification categorizes breast cancer either in “in situ” or “invasive”. In addition to this classification, it is crucial to assess the receptor status of a tumour, as it may determine the possibility of using targeted treatments in cancer therapy. On this basis, various subgroups can be identified, according to their profile of gene expression and positivity to oestrogen receptor (ER), progesterone receptor (PR) and the tyrosine kinase receptor, HER2. Tumours that lack expression of all three receptors are defined as Triple Negative Breast Cancer (TNBC). In recent years several studies on gene expression profiles have been conducted using high-throughput technologies, in order to identify molecular subtypes of breast cancer and to allow a better understanding of the complexity of the disease. Based on hierarchical clustering of gene expression microarrays, six subtypes of breast cancers have been identified: the luminal A, luminal B, HER2 positive subtype, the basal-like subtype, the normal breast subtype and the claudin-low subtype. The term retinoids refers to a group of compounds comprising metabolites and analogues of vitamin A, both natural and synthetic. The natural retinoids are essential components of diet and physiological regulators of many essential biological processes, such as embryonic development, metabolism and haematopoiesis. In adult mammals, retinoids such as All-Trans Retinoic Acid (ATRA), control homeostasis of different organs and tissues. All-trans retinoic acid (ATRA) is a small lipophilic molecule and an important regulator of gene expression. The biological action of ATRA and its derivatives is mediated by two classes of nuclear receptors for retinoids called Retinoic Acid Receptor (RAR) and Retinoic X Receptor (RXR). The receptors are ligand-dependent transcription factors that control the activity of several target genes either through a direct or indirect mechanism. Tumorigenesis is a multistep process characterized by a series of inherited or acquired genetic changes (mutations, chromosomal rearrangements, epigenetic phenomena), leading to a disruption of cellular homeostasis and development of the neoplastic process. Several lines of evidence indicate an important role of retinoids in homeostasis. It is clear that ATRA is able to act through different genomic mechanisms as well as to interact with other intracellular signalling systems, that provide the basis for its pleiotropic action. Currently, the best example of the anticancer action of retinoids is the use of ATRA in the treatment of patients suffering from acute promyelocytic leukaemia (APL). The retinoids have been investigated extensively for the prevention and treatment of cancer, predominantly because of their ability to induce cellular differentiation and to block proliferation. To evaluate the response of breast cancer cell lines to the anti-proliferative effect exerted by retinoic acid, Bolis and colleagues first defined the profile of ATRA-sensitivity in a panel of 48 breast cancer cell lines of the Cancer Cell Lines Encyclopedia (CCLE), comprising all subtypes of the disease. The drug response of each cell line has been quantified by computation of a sensitivity score (ATRA-score), scaled from 0 to 1. The Notch signaling pathway is an evolutionarily conserved regulator of cell fate, differentiation and growth in mammals: it is involved in homeostasis maintenance of tissues, including mammary gland. The Notch signaling is initiated by ligand binding. This triggers a series of proteolytic cleavage events, culminating in the release of the Notch intracellular domain (NICD). NICD is the active form of the receptor and translocates to the nucleus, acting as transcription factor of various gene families. According to the pleiotropic role of Notch in the development and maintenance of homeostasis, deregulation of Notch signaling is involved in hereditary diseases, haematological malignancies and solid tumours. Notch signaling has a role in physiological development of mammary gland and genetic alterations enhance breast cancer formation. Notch1 is oncogenic in 13% of triple negative breast cancers (TNBC). Among breast cancer subtypes, TNBCs are known to be the most aggressive, with pessimistic prognosis and with chemotherapy as the only therapeutic strategy available. Therefore, the need of new therapeutic strategies is evident. For tumours that over-express Notch1, γ-secretase inhibitor (GSI) is a potential element to block the Notch signaling. γ-secretase is the enzyme responsible for the release of Notch active form. In this study three TNBC cell lines, characterized by the higher ATRA-score among basal cell lines and by intragenic deletion of Notch1 gene, are studied: HCC1599 (ATRA-score =1), MB-157 (ATRA-score = 0.28) and MDA-MB-157 (ATRA-score = 0.24). A promising therapeutic approach for the most aggressive subtype of breast cancer is the combination of ATRA and of γ-secretase inhibitor. In this thesis I analyse RNA-sequencing data of TNBC cell lines treated for 8h with ATRA (1 μM), DAPT (1 μM) and their combination. The first phase of the study is focused on the definition of transcriptional changes of each drug. Each TNBC cell lines is analysed separately because the major variance between the samples is correlated to genotype and not to the treatment. By considering ATRA, the number of differentially expressed genes and the anti-proliferative effect are proportional to ATRA-score. Indeed, MDA-MB-157 with minor score is not sensitive to this drug: gene set enrichment analysis leads to none one significantly altered hallmark. In other cell lines, ATRA activates interferon α and γ signaling and inhibits cell growth. In MB-157 phase G1 of cell cycle is altered; in addition, mTOR signaling, involved in tumour progression, and Myc cascade are down-regulated in HCC1599. To better define the maximum effect of ATRA, GSEA of HCC1599 involves other gene sets as KEGG (Kyoto Encyclopedia of Genes and Genomes), REACTOME and GENE ONTOLOGY. In all results is evident that the administration of ATRA significantly reduces transcription of mitochondria, RNA polymerase I and II. Reduction of protein production, catabolites and ATP blocks the physiologic life of a cell, leading to cell apoptosis. In addition, what is discovered is that DAPT, γ-secretasi inhibitor, blocks the Notch targets and Myc cascade, since Myc is a direct target of Notch. The anti-proliferative effect of DAPT, in terms of down-regulation of proliferation-related hallmarks, results only in HCC1599 and in MB-157. Through viability assays, some members of Molecular Biology Laboratory of IRCCS Mario Negri measured cell growth of TNBC cell lines for 9 days. The viability of HCC1599 and MB-157 is more reduced than one of MDA-MB-157. It is consistent with GSEA results of proliferation reduction: time needed to reduce proliferation at transcription level is longer for MDA-MB-157. Furthermore, a study of Stoeck et al. suggested that breast cancer cell lines with NOTCH gene rearrangements that disrupt the NRR coding region were significantly more sensitive to GSI, in particular MRK-003. The Notch gene rearrangements is a necessary but not sufficient condition of sensitivity to GSI administration. It is related to the concentration of NICD: lower the concentration, lower the sensitivity of GSI is. They considered MDA-MB-157 a resistant cell line due to low NICD levels, despite the presence of intragenic deletion of Notch1. The Mario Negri laboratory data shows MDA-MB-157 with the lower NICD concentration and the lower cell growth inhibition. From the point of view of pharmacology, a drug combination may produce synergistic, additive, antagonistic or even suppressive effect if the combinational effect is greater than, equal to or less than the sum of each individual drug. To define the amount of anti-proliferative effect of the combination of ATRA e DAPT, isobologram analysis, visualization approach to evaluate the interaction between two drugs, has been performed. The synergy of ATRA and DAPT is validated only for HCC1599 and MB-157. For the second phase of the study, the aim is to define the molecular mechanism of synergistic effect between ATRA and DAPT based on sequencing data. The first approach is the comparison of differentially expressed genes by each compound in each cell lines. Unfortunately, there is no good result. Then, a function of DESeq2, R package for the differential expression analysis, is used. It is the interaction term and it demonstrates if the log2 fold change attributable to a given condition is different based on another factor. The interaction term can select only genes whose expression is changed by interaction of ATRA and DAPT in a manner more than additive with ATRA+DAPT. A list of 104 significantly genes, derived by analysis of all cell lines in a single workspace, are used to generate a protein-protein interaction network. Due to the absence of interesting connections between these genes and RARα, RARβ and RARγ and Notch1, targets of ATRA and DAPT respectively, differential expression analysis with interaction term is performed only with HCC1599 and MB-157, whose synergistic effect is validated. The 255 significantly genes are in PPI network with addition of targets of ATRA and DAPT. The two genes connecting RARα, RARβ and Notch1 are two: GATA2 and CCDN2 (cyclin D2). However, only two genes are not enough to define the molecular mechanism of synergy. The genes with interaction activity are obtained in each cell lines and, through a protein-protein interaction network, the neighbour community is investigated. We select only genes of neighbour community connected to al least two genes with interaction activity. Then, another filter is applied based on gene expression levels of these 825 genes. Only nodes with |log_2〖FC|>0.5 〗in one of the treatments are selected. A final network of 144 genes is obtained and over-representation pathway analysis is performed. The molecular mechanisms over-represented in the enrichment analysis are PI3k-AKT-mTOR signaling, TNF-α activation via NF-kB and Hedgehog signaling. Future experiments will allow us to better understand the molecular mechanism of the synergy of the combination. The interest in combination therapies derives from the fact that they show the therapeutic benefit and the clinical impact of therapy, thanks to the possibility of the modulation of multiple molecular targets simultaneously. Future studies will allow us to experimentally validate the outcomes of this work and in case of positive results it could lead to the definition of a new therapy for TNBC. Finally, for the availability of RNA-seq data of the same samples treated with ATRA for 24h, the thesis work focused on establishing the effect of ATRA on pathways over time. By analysing the two time points separately, the activation of interferon signaling and the antiproliferative effect are confirmed. The change, in term of amplification or reduction, is proportional to the time.
Il carcinoma della mammella è il tumore più frequentemente diagnosticato tra le donne nel mondo ed è la principale causa di morte per tumore. Il carcinoma della mammella è una malattia eterogenea che comprende entità distinte in termini di istologia, caratteristiche molecolari, prognosi clinica e risposta ai trattamenti. Questa diversità ha reso più complesso lo sviluppo di classificazioni clinicamente utili per determinare il comportamento di un tumore sulla base delle sue caratteristiche biologiche. La classificazione istopatologica si basa sulle differenti caratteristiche morfologiche dei tumori e classifica il tumore in in situ o invasivo. Negli ultimi decenni è stata dimostrata la fondamentale importanza di due recettori di ormoni steroidei, il recettore degli estrogeni (ER) e del progesterone (PR), e del recettore tirosin-chinasico HER2 (human epidermal growth factor receptor 2) per l’eziologia, la prognosi e la terapia dei tumori della mammella. Accanto alla classificazione istopatologica, i carcinomi al seno si possono distinguere sulla base dell’espressione dei suddetti recettori, valutata mediante analisi immunoistochimica, che permette di rilevare la presenza della proteina. L’analisi dell’espressione genica, resa possibile dallo sviluppo di tecniche basate su microarray a cDNA, ha permesso di suddividere i tumori in diversi sottotipi molecolari sulla base della somiglianza del profilo di espressione genica. In questo modo è stata definita una classificazione, che ha individuato sei diversi sottotipi. Con il termine retinoidi ci si riferisce a tutte le molecole strutturalmente e funzionalmente analoghe al retinolo (Vitamina A), sia naturali che sintetiche. La vitamina A e i suoi metaboliti biologicamente attivi sono molecole essenziali per lo sviluppo embrionale, il meccanismo della visione e l’omeostasi di numerosi tessuti e sistemi, tra cui il sistema nervoso, immunitario e riproduttivo. ATRA è una piccola vitamina liposolubile, importante regolatrice dell’espressione genica. L’acido retinoico e i suoi derivati regolano infatti l’espressione di geni coinvolti nella crescita e nel differenziamento cellulare attraverso specifici recettori nucleari, RAR (retinoic acid receptor) e RXR (retinoid X receptor). I retinoidi svolgono un ruolo importante nel mantenimento dell’omeostasi e quindi posso essere correlati alla carcinogenesi. La perdita della loro attività o la diminuzione dei loro livelli intracellulari è associata ad una crescita cellulare aberrante e allo sviluppo di un’ampia varietà di tumori. Inoltre, l’acido retinoico tutto-trans è stato il primo agente anti-proliferativo e cito-differenziante ad essere utilizzato in clinica nel trattamento di un raro sottotipo di leucemia mieloide acuta (AML, acute myeloid leukemia), la leucemia promielocitica acuta (APL, acute promyelocytic leukemia). Grazie al successo di ATRA nell’ambito della leucemia promielocitica acuta, l’interesse per il potenziale utilizzo terapeutico dei retinoidi si è esteso anche ad altri tipi di carcinomi, come il tumore della mammella. Per valutare la risposta delle linee cellulari di carcinoma mammario all'effetto antiproliferativo esercitato dall'acido retinoico, Bolis et al. hanno definito il profilo di sensibilità ad ATRA in un pannello di 48 linee cellulari di carcinoma mammario del CCLE (Cancer Cell Lines Encyclopedia). La risposta farmacologica di ciascuna linea cellulare è stata quantificata per il calcolo finale di un punteggio di sensibilità (ATRA-score), che varia da 0 ad un massimo di 1. La via di trasduzione del segnale di Notch fisiologicamente è coinvolta nello sviluppo embrionale e nel mantenimento dell’omeostasi dei tessuti adulti, tra cui la ghiandola mammaria, attraverso la regolazione di proliferazione, differenziamento e apoptosi. La trasduzione del segnale dunque inizia con l’interazione tra Notch e il suo ligando espresso su una cellula adiacente, o eventualmente sulla stessa cellula. Il legame che si forma innesca una serie di eventi proteolitici che portano al rilascio della porzione intracellulare attiva del recettore, NICD (Notch Intracellular Domain). NICD è la forma attiva del recettore che trasloca nel nucleo dove regola la trascrizione dei suoi geni bersaglio. Dato il ruolo pleiotropico di Notch nello sviluppo e nel mantenimento dell’omeostasi di diversi organi e tessuti, una deregolazione della sua via di trasduzione del segnale è associata a numerose patologie umane, tra cui malattie ereditarie, tumori ematologici e tumori solidi. Notch1 ha un ruolo oncogenico nel 13% dei tumori della mammella tripli negativi. Uno dei potenziali bersagli indagati per i tumori sovra-esprimenti Notch1 è rappresentato dalla γ-secretasi, l’enzima che catalizza il taglio proteolico che rilascia la forma attiva di Notch nel citoplasma. In questo particolare lavoro sono state studiate tre linee cellulare TNBC aventi ATRA-score maggiore per il fenotipo basale e caratterizzate dalla delezione intragenica del gene Notch1: HCC1599 (ATRA-score =1), MB-157 (ATRA-score = 0.28) e MDA-MB-157 (ATRA-score = 0.24). Si è concentrato sul determinare le perturbazioni trascrittomiche in risposta al trattamento per 8h con ATRA (1 μM), DAPT (1 μM), inibitore di γ secretasi, e con la combinazione dei due in linee cellulari di TNBC, attraverso l’analisi di dati di sequenziamento del RNA. La prima parte del lavoro di tesi, si è focalizzata sul comprendere le perturbazioni trascrittomiche indotte dai singoli agenti. Per fare ciò, poiché la maggior varianza tra i campioni è dovuta alla linea e non al trattamento, ciascun tipo cellulare è stato analizzato separatamente. Considerando l’acido retinoico tutto-trans, la risposta di ciascuna linea cellulare risulta proporzionale all’ATRA-score in base al numero di geni differenzialmente espressi e all’entità dell’effetto antiproliferativo. Infatti, MDA-MB-157, con score minimo di 0.25, hanno mostrato una scarsa sensibilità al farmaco. Nei restanti tipi cellulari, ATRA attiva il pathway dell’interferone α e γ e inibisce la crescita cellulare. Cercando di caratterizzate l’effetto massimo di ATRA, esclusivamente per le HCC1599 sono stati analizzati i gene set di KEGG (Kyoto Encyclopedia of Genes and Genomes), REACTOME e GENE ONTOLOGY. Dai risultati dei gene set significativamente alterati, ATRA abbassa la trascrizione dei mitocondri, dell’importazione di proteine nel mitocondrio stesso e dell’RNA polimerasi I e II, così da impedire l’accesso al promotore, e quindi la trascrizione. D’altra parte, l’inibitore di γ secretasi, noto con il nome DAPT, interferisce con i processi molecolari attivati dal gene Notch1. In questo lavoro di tesi, il risultato comune a seguito della somministrazione del composto è un arresto della trascrizione dei target di Notch, tra cui il fattore di trascrizione Myc e, dunque, la sua cascata. Concentrandosi sull’inibizione della crescita cellulare, l’analisi della trascrittomica tramite RNA-Seq conferma una deregolazione della crescita cellulare rispettivamente per HCC1599 e MB-157. Quando si combinano due o più farmaci insieme si può avere sinergia, un effetto additivo, un antagonismo o un effetto repressivo in base a se la combinazione ha effetti maggiori, uguali o minori della somma degli effetti dei composti individualmente. Al fine di comprendere se l’aumento dell’attività anti-proliferativa data in caso di combinazione del farmaco rispetto ai singoli composti sia dovuto ad un’azione additiva o sinergica degli stessi, è stato applicato il metodo degli isobologrammi. Per HCC1599 e MB-157 è dimostrata la sinergia. Con lo scopo di determinare il meccanismo molecolare alla base della sinergia tra ATRA e DAPT dai dati di RNA-seq, sono state affrontate indagini di diversa natura. Il primo approccio si è basato sulla comparazione dei risultati, in termini di geni, ottenuti dall’analisi del trattamento di ATRA e DAPT, senza ottenere risultati. Le indagini seguenti hanno utilizzato una caratteristica della funzione DESeq2, usata per l’analisi di espressione differenziale. Nella prima ricerca, la selezione dei geni con regolazione sinergica è stata eseguita considerando i tre tipi cellulari contemporaneamente e visualizzando in un network d’interazione proteina-proteina come essi interagissero con i target diretti di ATRA e DAPT. L’assenza di risultato ha portato alla ri-analisi dei dati con l’eliminazione delle MDA-MB-157, tipo cellulare più resistente ad entrambi i composti. In questo caso, nel PPI network i punti di comunicazione tra RARα, RARβ e RARγ, target dell’acido retinoico tutto-trans, e Notch1, target di DAPT, sono due geni dei 255 statisticamente significativi: GATA2 e CCDN2 (ciclina D2). Nuovamente, il trovare due geni non ha permesso di identificare il meccanismo molecolare alla base della sinergia. Alla luce di questi esiti, sono stati ricavati i geni con attività d’interazione in ciascun sottotipo cellulare e, tramite un network d’interazione proteina-proteina, è stata indagata la “neighbour community”. All’interno di essa, sono stati identificati gli elementi che interagiscono con almeno due dei geni con regolazione sinergica e che avessero un cambiamento in termini di log_2〖FC 〗in modulo maggiore di 0.5 in una delle terapie. Con un’analisi di over-rappresentazione, i pathways molecolari che rappresentano i 144 geni finali sono la cascata PI3K-AKT-mTOR, l’attivazione di NF-kB indotta da TNF-α e il signaling di Hedgehog. Esperimenti futuri permetteranno di comprendere maggiormente la sinergia della combinazione. L’interesse nelle terapie combinante nasce dal fatto che esse mostrano un successo dal punto di vista clinico. L’introduzione di terapie combinate ha lo scopo di aumentare il beneficio terapeutico e l’impatto clinico della terapia in quando si modulano più target molecolari contemporaneamente. Studi futuri permetteranno di validare sperimentalmente quanto trovato in questo lavoro e, se gli esiti sono positivi, definire una nuova possibile terapia per il carcinoma alla mammella triplo negativo. Infine, per la disponibilità dei dati di RNA-seq dei medesimi campioni trattati con l’acido retinoico tutto-trans per 24h, il lavoro di tesi si è focalizzato sullo stabilire le ripercussioni che ATRA ha sui pathway nel tempo. Analizzando i due time point separatamente, l’attivazione del signaling dell’interferone e l’effetto antiproliferativo sono stati confermati. L’entità del cambiamento, intesa come amplificazione o riduzione, si è dimostrata crescere proporzionalmente al passare del tempo.
Terapia combinata per il carcinoma mammario : analisi di dati di trascrittomica dai singoli composti allo studio della sinergia
TROIANI, MARTINA
2017/2018
Abstract
Among females, breast cancer is the most common diagnosed cancer and the leading cause of tumour death. In the last decade, it is notable a decreasing trend thanks to presence of primary prevention programs, of diagnostic screening (mammogram) and of new therapeutic strategies. At present, the broad heterogeneity of breast cancer reflects the well-accepted notion that there is not just one disease with disparate variant subtypes, but that breast cancer instead represents a collection of distinct neoplastic diseases of the breast and the cells composing it. Behind this complexity, several systems have been developed to classify this very highly heterogeneous disease and possibly to get information about tumour behaviour and provide more effective therapies. Histological classification categorizes breast cancer either in “in situ” or “invasive”. In addition to this classification, it is crucial to assess the receptor status of a tumour, as it may determine the possibility of using targeted treatments in cancer therapy. On this basis, various subgroups can be identified, according to their profile of gene expression and positivity to oestrogen receptor (ER), progesterone receptor (PR) and the tyrosine kinase receptor, HER2. Tumours that lack expression of all three receptors are defined as Triple Negative Breast Cancer (TNBC). In recent years several studies on gene expression profiles have been conducted using high-throughput technologies, in order to identify molecular subtypes of breast cancer and to allow a better understanding of the complexity of the disease. Based on hierarchical clustering of gene expression microarrays, six subtypes of breast cancers have been identified: the luminal A, luminal B, HER2 positive subtype, the basal-like subtype, the normal breast subtype and the claudin-low subtype. The term retinoids refers to a group of compounds comprising metabolites and analogues of vitamin A, both natural and synthetic. The natural retinoids are essential components of diet and physiological regulators of many essential biological processes, such as embryonic development, metabolism and haematopoiesis. In adult mammals, retinoids such as All-Trans Retinoic Acid (ATRA), control homeostasis of different organs and tissues. All-trans retinoic acid (ATRA) is a small lipophilic molecule and an important regulator of gene expression. The biological action of ATRA and its derivatives is mediated by two classes of nuclear receptors for retinoids called Retinoic Acid Receptor (RAR) and Retinoic X Receptor (RXR). The receptors are ligand-dependent transcription factors that control the activity of several target genes either through a direct or indirect mechanism. Tumorigenesis is a multistep process characterized by a series of inherited or acquired genetic changes (mutations, chromosomal rearrangements, epigenetic phenomena), leading to a disruption of cellular homeostasis and development of the neoplastic process. Several lines of evidence indicate an important role of retinoids in homeostasis. It is clear that ATRA is able to act through different genomic mechanisms as well as to interact with other intracellular signalling systems, that provide the basis for its pleiotropic action. Currently, the best example of the anticancer action of retinoids is the use of ATRA in the treatment of patients suffering from acute promyelocytic leukaemia (APL). The retinoids have been investigated extensively for the prevention and treatment of cancer, predominantly because of their ability to induce cellular differentiation and to block proliferation. To evaluate the response of breast cancer cell lines to the anti-proliferative effect exerted by retinoic acid, Bolis and colleagues first defined the profile of ATRA-sensitivity in a panel of 48 breast cancer cell lines of the Cancer Cell Lines Encyclopedia (CCLE), comprising all subtypes of the disease. The drug response of each cell line has been quantified by computation of a sensitivity score (ATRA-score), scaled from 0 to 1. The Notch signaling pathway is an evolutionarily conserved regulator of cell fate, differentiation and growth in mammals: it is involved in homeostasis maintenance of tissues, including mammary gland. The Notch signaling is initiated by ligand binding. This triggers a series of proteolytic cleavage events, culminating in the release of the Notch intracellular domain (NICD). NICD is the active form of the receptor and translocates to the nucleus, acting as transcription factor of various gene families. According to the pleiotropic role of Notch in the development and maintenance of homeostasis, deregulation of Notch signaling is involved in hereditary diseases, haematological malignancies and solid tumours. Notch signaling has a role in physiological development of mammary gland and genetic alterations enhance breast cancer formation. Notch1 is oncogenic in 13% of triple negative breast cancers (TNBC). Among breast cancer subtypes, TNBCs are known to be the most aggressive, with pessimistic prognosis and with chemotherapy as the only therapeutic strategy available. Therefore, the need of new therapeutic strategies is evident. For tumours that over-express Notch1, γ-secretase inhibitor (GSI) is a potential element to block the Notch signaling. γ-secretase is the enzyme responsible for the release of Notch active form. In this study three TNBC cell lines, characterized by the higher ATRA-score among basal cell lines and by intragenic deletion of Notch1 gene, are studied: HCC1599 (ATRA-score =1), MB-157 (ATRA-score = 0.28) and MDA-MB-157 (ATRA-score = 0.24). A promising therapeutic approach for the most aggressive subtype of breast cancer is the combination of ATRA and of γ-secretase inhibitor. In this thesis I analyse RNA-sequencing data of TNBC cell lines treated for 8h with ATRA (1 μM), DAPT (1 μM) and their combination. The first phase of the study is focused on the definition of transcriptional changes of each drug. Each TNBC cell lines is analysed separately because the major variance between the samples is correlated to genotype and not to the treatment. By considering ATRA, the number of differentially expressed genes and the anti-proliferative effect are proportional to ATRA-score. Indeed, MDA-MB-157 with minor score is not sensitive to this drug: gene set enrichment analysis leads to none one significantly altered hallmark. In other cell lines, ATRA activates interferon α and γ signaling and inhibits cell growth. In MB-157 phase G1 of cell cycle is altered; in addition, mTOR signaling, involved in tumour progression, and Myc cascade are down-regulated in HCC1599. To better define the maximum effect of ATRA, GSEA of HCC1599 involves other gene sets as KEGG (Kyoto Encyclopedia of Genes and Genomes), REACTOME and GENE ONTOLOGY. In all results is evident that the administration of ATRA significantly reduces transcription of mitochondria, RNA polymerase I and II. Reduction of protein production, catabolites and ATP blocks the physiologic life of a cell, leading to cell apoptosis. In addition, what is discovered is that DAPT, γ-secretasi inhibitor, blocks the Notch targets and Myc cascade, since Myc is a direct target of Notch. The anti-proliferative effect of DAPT, in terms of down-regulation of proliferation-related hallmarks, results only in HCC1599 and in MB-157. Through viability assays, some members of Molecular Biology Laboratory of IRCCS Mario Negri measured cell growth of TNBC cell lines for 9 days. The viability of HCC1599 and MB-157 is more reduced than one of MDA-MB-157. It is consistent with GSEA results of proliferation reduction: time needed to reduce proliferation at transcription level is longer for MDA-MB-157. Furthermore, a study of Stoeck et al. suggested that breast cancer cell lines with NOTCH gene rearrangements that disrupt the NRR coding region were significantly more sensitive to GSI, in particular MRK-003. The Notch gene rearrangements is a necessary but not sufficient condition of sensitivity to GSI administration. It is related to the concentration of NICD: lower the concentration, lower the sensitivity of GSI is. They considered MDA-MB-157 a resistant cell line due to low NICD levels, despite the presence of intragenic deletion of Notch1. The Mario Negri laboratory data shows MDA-MB-157 with the lower NICD concentration and the lower cell growth inhibition. From the point of view of pharmacology, a drug combination may produce synergistic, additive, antagonistic or even suppressive effect if the combinational effect is greater than, equal to or less than the sum of each individual drug. To define the amount of anti-proliferative effect of the combination of ATRA e DAPT, isobologram analysis, visualization approach to evaluate the interaction between two drugs, has been performed. The synergy of ATRA and DAPT is validated only for HCC1599 and MB-157. For the second phase of the study, the aim is to define the molecular mechanism of synergistic effect between ATRA and DAPT based on sequencing data. The first approach is the comparison of differentially expressed genes by each compound in each cell lines. Unfortunately, there is no good result. Then, a function of DESeq2, R package for the differential expression analysis, is used. It is the interaction term and it demonstrates if the log2 fold change attributable to a given condition is different based on another factor. The interaction term can select only genes whose expression is changed by interaction of ATRA and DAPT in a manner more than additive with ATRA+DAPT. A list of 104 significantly genes, derived by analysis of all cell lines in a single workspace, are used to generate a protein-protein interaction network. Due to the absence of interesting connections between these genes and RARα, RARβ and RARγ and Notch1, targets of ATRA and DAPT respectively, differential expression analysis with interaction term is performed only with HCC1599 and MB-157, whose synergistic effect is validated. The 255 significantly genes are in PPI network with addition of targets of ATRA and DAPT. The two genes connecting RARα, RARβ and Notch1 are two: GATA2 and CCDN2 (cyclin D2). However, only two genes are not enough to define the molecular mechanism of synergy. The genes with interaction activity are obtained in each cell lines and, through a protein-protein interaction network, the neighbour community is investigated. We select only genes of neighbour community connected to al least two genes with interaction activity. Then, another filter is applied based on gene expression levels of these 825 genes. Only nodes with |log_2〖FC|>0.5 〗in one of the treatments are selected. A final network of 144 genes is obtained and over-representation pathway analysis is performed. The molecular mechanisms over-represented in the enrichment analysis are PI3k-AKT-mTOR signaling, TNF-α activation via NF-kB and Hedgehog signaling. Future experiments will allow us to better understand the molecular mechanism of the synergy of the combination. The interest in combination therapies derives from the fact that they show the therapeutic benefit and the clinical impact of therapy, thanks to the possibility of the modulation of multiple molecular targets simultaneously. Future studies will allow us to experimentally validate the outcomes of this work and in case of positive results it could lead to the definition of a new therapy for TNBC. Finally, for the availability of RNA-seq data of the same samples treated with ATRA for 24h, the thesis work focused on establishing the effect of ATRA on pathways over time. By analysing the two time points separately, the activation of interferon signaling and the antiproliferative effect are confirmed. The change, in term of amplification or reduction, is proportional to the time.File | Dimensione | Formato | |
---|---|---|---|
2019_04_Troiani.pdf
solo utenti autorizzati dal 05/04/2022
Descrizione: Testo della tesi
Dimensione
4.35 MB
Formato
Adobe PDF
|
4.35 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/146147