Splicing alterations in low risk myelodysplastic syndromes : TGF-B pathway deregulation in SF3B1 mutated patients

Myelodysplastic Syndromes (MDS) are hematological disorders marked by the abnormal growth of hematopoietic stem cells (HSC) with recurring genetic anomalies, dysplaStic changes in myeloid precursors, inefficient blood cell production, low blood cell counts, and a substantial likelihood of progression to acute myeloid leukemia (AML). Typically, these conditions affect older individuals and can arise spontaneously (de novo) or result from previous exposure to cancer-causing agents or therapies (therapy-related MDS). MDS subtypes are determined by factors such as the extent of dysplasia in different blood cell lineages, the proportion of blasts in both bone marrow (BM) and peripheral blood (PB), the presence of ring sideroblasts, as well as the specific genetic abnormalities. Gene mutations are a common occurrence in MDS patients. They impact many genes and biological pathways, including the splicing of RNA, the regulation of epigenetic processes, transcription factors etc. Among the most frequently mutated genes is the splicing pathway, with frequent mutations in SF3B1, SRSF2, U2AF1 and ZRSR2. These mutations in MDS disrupt the normal functioning of splicing factors (SF), leading to alterations in RNA processing. Consequently, this leads to the generation of RNA variants with altered or compromised biological functions, ultimately resulting in shifts in aberrant alternative splicing and gene expression patterns. In the first part of this dissertation, we analyzed the landscape and gene expression patterns in patients with MDS, focusing on genes related to RNA splicing. The study group consisted of 50 MDS patients, including 30 males and 20 females, with a median age of 75 years at the time of diagnosis (ranging from 40 to 91 years). These patients were selected from an initial cohort of 300 individuals, all of whom had paired DNA and RNA samples collected from their bone marrow mononuclear cells (BM-MNC) at the time of MDS diagnosis. More specifically, 32 out of the 50 patients were chosen using targeted-Next Generation Sequencing (t- NGS) based on the presence of mutations in splicing genes. The remaining 18 patients had an unknown mutational profile. As a control group, were used BM samples obtained from 64 patients without hematologic disorders. To identify possible co-mutations in association with splicing factors genes (SF3B1, SRSF2 and U2AF1), we conducted a comprehensive Whole Genome Sequencing (WGS) analysis on the entire study cohort of 50 MDS patients. The results from WGS revealed that FAM20C was among the most frequently mutated genes in patients with SF3B1, SRSF2 and U2AF1 mutations (70%, 50% and 71%, respectively) and KDM4B was among the most frequently mutated genes in patients with SRSF2 and U2AF1 mutations (50 % and 71 %, respectively), although the real pathogenicity of these somatic variants need to be better characterized. Moreover, the ACLY gene also showed variable mutation frequencies: it was frequently mutated in patients with SF3B1 mutations (90%) and SRSF2 mutations (50%), but not in patients with U2AF1 mutations (0%). Data obtained from transcriptome analysis using RNA sequencing (RNA-seq) revealed a significant alteration in the expression profile of MDS patients (n=46) compared to control subjects (n=64). Three hundred genes exhibited increased expression (up-regulated), while 3921 genes displayed decreased expression (down-regulated). Further analysis, considering the presence of mutations in SF3B1, SRSF2, and U2AF1, allowed the identification of differentially expressed genes (DEG) within each patient subgroup. Notably, 622 DEGs were found in patients with SF3B1 mutations, 13 DEGs were identified in those with SRSF2 mutations, and only 4 genes exhibited differential expression in patients with U2AF1 mutations. DEGs, which were selected based on specific criteria (padj < 0.05 and log2FC ≥ 2 and ≤ - 2), were used to conduct a gene ontology using the DAVID bioinformatics tool. This analysis revealed a total of 37 pathways with disrupted regulation in MDS patients who had SF3B1 mutations. In contrast, no pathways were identified as significantly altered in patients with SRSF2 and U2AF1 mutations. Subsequently, we chose to focus on the TGF-β pathway in patients with SF3B1 mutation for further investigation. This choice was influenced also by the recent approval of a novel drug, a recombinant protein known as Luspatercept. This has the capability to modulate signal transduction by targeting the TGF-β pathway, with the aim of restoring normal erythropoiesis, particularly in Low-Risk MDS patients experiencing ineffective hematopoiesis. We found 6 DEGs associated with the TGF-β pathway (CHRD, DCN, SMAD9, FST, ID4, PITX2) and 10 DEGs known to influence TGF-β activity (BDNF, BMPER, COL1A1, COL1A2, FN1, GLI2, IGF1, MMP2, NFIB, and PLAU). The expression of these genes (DCN, FST, SMAD9, FN1, IGF1, MMP2 and PLAU) was validated by quantitative real-time PCR (Q-RT-PCR), confirming the observed gene deregulation, which had initially been identified through RNA-Seq analysis. More specifically, we observed a significant increase in the mRNA expression levels of DCN, FST, SMAD9, FN1, IGF1, MMP2, and a noteworthy decrease in the mRNA expression level of PLAU in patients with SF3B1 mutations, in comparison to patients without splicing factor mutations and to BM-MNCs isolated from healthy donors. Currently, we are in the process of gathering a new group of MDS patients, including those with SF3B1 mutations and those without SF mutations (referred to as SF wild-type), as part of a prospective study. The aim is to confirm our findings in an independent cohort of patients. In the second part of this dissertation, we focused on alternative splicing (AS). Frequently, a single pre-mRNA can give rise to multiple distinct mature mRNAs through a mechanism known as AS. More than 90% of human genes experience AS, resulting in the generation of ten or more isoforms from a single expressed gene on numerous occasions. The analysis of RNA sequencing data was extended to delve into the study of AS. The percentage of genes regulation through AS exhibits considerable variation and was evident in all patient groups with splicing factor (SF) mutations (SF3B1, SRSF2 and U2AF1). Particularly, AS was most prominent in patients with SF3B1 mutations, impacting approximately 25.70% of genes. SRSF2 mutations were linked to AS in 20.92% of genes, and U2AF1 mutations showed AS regulation in around 13.15% of genes. More specifically, in SF3B1 K700E, a wide array of genes exhibited AS events, with "Exon Skipping" (ES) and "Mutually Exclusive Exons" (MXE) being the most prevalent types. They constituted 28% and 25 % of the total AS events, respectively. Conversely, in the SRSF2 and U2AF1 mutated groups, ES events were more conspicuous, representing over half of the AS events at 54% and 57%, respectively. MXE events followed at 9% and 11%, respectively. Following the characterization of AS within the three categories, our analysis extended to comparing these AS events between SF3B1 K700E vs. SRSF2 mutated patients to identify commonalities and differences in AS patterns across this category. We identified many genes that exhibited common regulation, whether they were upregulated or downregulated in the same or different directions (which will be discussed in more detail in Results). Last but not least, to comprehend the biological relevance of the identified AS events, we conducted a Gene Ontology (GO) analysis. Functional annotations identified iron metabolism, heme and coenzyme A synthesis among the most deregulated pathways. Of note, SF3B1 mutated patients showed AS in genes belonging to TGF-β pathway (TGFBR1, TGFBR2, TGFBR3, MAP3K7, FKBP12, FKBP1A etc.), suggesting that not only differentially expressed but also aberrant spliced genes may have a role in the pathogenesis of SF3B1 mutated MDS patients. In summary, this dissertation confirms that MDS patients exhibit altered and highly diverse gene expression profiles. This diversity may be linked to specific mutations in genes related to the splicing process. Moreover, these patients display a distinct form of aberrant splicing, shedding light on the complex molecular landscape at play.