Long Non-coding RNAs and Their Role on Epigenome as Diagnostic Markers in Childhood Acute Lymphoblastic Leukemia of T Cells.

Specific Aim 1 Evaluation of lncRNAs in a cohort of T-ALL patients in comparison with T-Naive lymphocytes and PBMC from healthy subjects. Identification of the best suitable in vitro models. To achieve the goal of Aim 1 we planned to enroll a cohort of pediatric patients affected by T-ALL at the Pausilipon hospital, the UO2, which is a reference center for the treatment of oncological diseases in Campania. For patient recruitment, specific and detailed informed consent will be used according to the current General Data Protection Regulation. The collected and processed biological samples will be stored in the SDN Biobank, which is the institutional biobank of IRCCS Synlab SDN and is a member of the BBMRI-ERIC infrastructure. Considering that childhood T-ALL is rare cancer, with an incidence in the pediatric population of 1 case every 100000 children, we plan to enroll 16-20 T-ALL patients. Starting from RNA-seq data previously obtained by our laboratory, comparing 6 pediatric T-ALL with naïve T-cells derived from cord blood, we identified a set of differentially regulated lncRNAs. From these, we selected 6 among the most up-regulated lncRNAs that have never been associated with leukemia, in general, and pediatric acute T-cell leukemia in particular. The 6 up-regulated selected lncRNAs are: AC002454.1 (ENSG00000237819), PCAT18 (ENSG00000265369), HHIP-AS1 (ENSG00000248890), LINC01222 (ENSG00000233410), AC116351.1 (ENSG00000215246), AC247036.1 (ENSG00000271201). Using ours and other public datasets of RNA-seq experiments, we will confirm our suggestions in silico, and then proceed to subsequent analyses on biological samples. The expression levels of these lncRNAs will be analyzed ex vivo, using RT-PCR experiments, in the selected cohort in comparison with PBMCs derived from healthy subjects and pediatric Acute Lymphoblastic Leukemia of B-cells (B-ALL) patients (already present in our Biobank) to ascertain over-expression in the T-ALL population and specificity with respect to another pediatric hematological disease such as B-ALL. Moreover, the transcript levels of the selected lncRNAs will be analyzed in a minimum of 5 cell lines model disease, for both T-ALL and B-ALL. These data will allow us to select the best cellular models for subsequent functional studies. The lncRNA expression data will be cross-referenced with clinical data from T-ALL pediatric subjects in order to identify a correlation between the expression of the lncRNAs and patient outcomes and to obtain a molecular signature of lncRNAs specific for T-ALL that will be useful for the diagnosis and for the improvement of pediatric patient management.

Specific Aim 2 Identification of cis-regulatory elements differentially accessible between pediatric T-ALL patients and healthy subjects, evaluation of epigenetic modifications of the loci identified above, and gene regulatory networks modeling to unravel the functional role of candidate lncRNAs in childhood T-ALL. Current advances in high-throughput sequencing and bioinformatic approaches highlighted that lncRNAs can regulate gene expression at the epigenetical, transcriptional and post-transcriptional levels. As for the epigenetic side, they can drive significant changes in chromatin architecture, DNA methylation and histone modifications. For example, lncRNAs may guide transcriptional regulators to specific genomic loci or facilitate the enhancer-promoter communication by operating a direct or indirect remodeling of the chromatin folding. Today it is well known that these kinds of epigenetic regulation mechanisms have a critical impact on different steps of malignancy onset and progression, from uncontrolled proliferation to apoptosis resistance, altering the chromatin accessibility and methylation state of cis-regulatory elements and therefore leading to aberrant gene expression in tumors with respect to the healthy counterpart. Despite the fact that epigenetic alterations are known to be a key factor in leukemia as well, scarce advances have been made to elucidate the pivotal role of lncRNAs on the epigenomic landscape of ALL patients, let alone pediatric ones. Only few specific lncRNAs have been identified and most of the questions about their regulatory function as potential oncogenes or tumor-suppressors remain unanswered and even unasked. For the Aim 2, we will unveil the regulatory dynamics interplaying between lncRNAs, epigenetic factors and oncogene expression. We plan to couple RNA-seq with ATAC-seq and MethylC-seq experiments in T-ALL pediatric patient samples in comparison with naïve T-cells from cord blood. ATAC-seq captures open chromatin regions, usually trimethylated at H3K4, H3K36, and H3K79, which typically correlate with cis-regulatory elements. ATACseq of T-ALL samples will provide genomic regions of interests to predict putative lncRNA-genome DNA binding sites via computational methods and, in combination with transcriptomics and functional analyses, this technology will allow us to bioinformatically model the dynamic interactions between genomic cis-regulatory elements and trans-effectors such as transcription factors, chromatin remodeling complexes and lncRNAs. Further, as emerging evidences have uncovered a crosstalk also between lncRNAs and DNA methylation, we will perform a genome-wide identification of cytosine DNA methylation states at single-base resolution through MethylC-seq technique in order to complement and enhance our gene regulatory network model. Once the epigenetic landscapes of T-ALL patients have been defined, we intend to characterize also the chromatin and transcriptomic landscape in T-ALL cell line models, both in control conditions and upon the knockdown of the target lncRNAs (see Aim3 for further details). This experiment would help us to corroborate and sharpen our model of the operating mechanism for lncRNAs under investigation. This cross-firing strategy of functional experiments and computational models will provide mechanistic understanding of epigenetic and transcriptomic relations in pediatric TALL and, above all, will pave the way to novel strategies for diagnostic and therapeutic intervention.

Specific Aim 3 Evaluation, using appropriate T-ALL cell line model systems, of the role of identified lncRNAs on epigenome modifications.

Once evaluated the epigenome of T-ALL patients in comparison with healthy subjects, we will proceed to assess the involvement of identified lncRNAs in leukemogenesis. The activities of Aim3 will be conducted in collaboration with the UO3, which has all the tools and skills to support this experimental phase. We will proceed to the transient silencing of the lncRNAs identified, in the T-ALL model cell lines that express them at the highest levels. For the silencing experiments, we will use traditional (lipofectamine or lentiviral vectors) or innovative methods. In particular, a method recently developed by an American company (FANA-ASO, oligonucleotide, AUMBiotech, USA) based on the use of modified oligonucleotides able to enter into hematopoietic cells without the use of transfection agents, known to be toxic to cells. The FANA-ASO is a technology abundantly used for the silencing of target transcripts in hematopoietic models that are difficult to transfect using classical transfection methods. They also provide high silencing efficiency for molecules such as lncRNAs.

In the selected T-ALL model systems, each lncRNA will be silenced individually. After the transient transfection, the efficiency of the silencing will be verified by q-RT-PCR at different time points. Then, we propose to analyze the effects of lncRNA knockdown on tumor progression by evaluating changes in biological behaviors of the transfected cells (i.e.

proliferation, cell cycle progression, and motility abilities), changes in cell phenotype (by performing three-dimensional culture assay), epithelial-mesenchymal transition (EMT), stem cell markers, oxidative stress, and drug sensitivity (by performing viability assay or IC50 calculation). Depending on the results obtained, coupled silencing experiments with two lncRNAs simultaneously will also be considered. For the lncRNAs that will have had the greatest effect on the selected models, we will evaluate what are the effects that lncRNAs have on chromatin reorganization, through the application of the NGS methods reported in Aim 2, comparing silenced lncRNA models respect to the untreated control. These data will allow us to evaluate the relationships between the expression levels of identified lncRNAs and peculiar chromatin alterations in leukemic T cells. These experiments prompted us to identify a specific gene regulatory network associated with the selected lncRNAs. Passing from the laboratory bench to the patient beds, we will silence the selected lncRNAs directly in the childhood T-Blasts and we will evaluate the variations in the levels of the gene identified as part of the gene regulatory network of the selected lncRNA.

The identified alterations could be then correlated with the clinical features of the disease in order to evaluate their impact on therapeutic response.