The effects of SF1 overexpression and its cancer-related mutations on SRSF2 subcellular localization, expression, and global splicing regulation

• Ano de defesa: 2025
• Autor(a) principal: Sena, Mariana de Souza
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Splicing factor 1; Splicing; Immunofluorescence; Imunofluorescência; Mutações; Mutations; SRSF2
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/17/17136/tde-03112025-172853/

Splicing is the cellular process that produces mature mRNAs and is carried out by the spliceosome. Defects in splicing produce aberrant transcripts. The identification of somatic mutations in spliceosome genes has implicated the splicing machinery in cancer development. SF1 (Splicing factor 1) is a protein that acts in the early stages of spliceosome assembly and some studies report SF1 mutations in hematologic malignancies as well as SF1 differential expression in some types of cancer. In silico analyses from our group identified the occurrence of SF1 mutations in various types of cancer and that most of them were classified as harmful with high reliability. SRSF2 is another splicing factor mostly mutated in cancer and its pre mRNA is bound by SF1. However, the impact of aberrant SF1 activity on SRSF2 function and its consequences for the splicing process remains poorly understood. Thus, the objective of this study was to assess how the overexpression of SF1 and the presence of different mutations in this protein affects the subcellular localization of SRSF2 and the splicing process. The subcellular localization of SRSF2 was analyzed by immunofluorescence assay in HeLa cells transiently overexpressing the wild-type and mutated forms of SF1, namely SF1K64N, SF1R102H, SF1G181R, SF1R255W, SF1W265L, SF1R345C, SF1G372V and SF1T454M. The localization pattern of SRSF2 yields different phenotypes, which were counted and quantified by the ImageJ software. To better characterize the cytoplasmic and nuclear granules containing SRSF2, Z-stack images were acquired using a confocal microscope. HeLa cells were transduced with an empty vector (MIG), MIG SF1 WT and mutated SF1 (SF1 G372V and SF1 T454M) to assess SRSF2 subcellular localization under stably overexpression conditions. SRSF2 fluorescence intensity was quantified using corrected total cell fluorescence (CTCF). Moreover, a splicing assay using the pTN24 reporter plasmid was performed with these transduced cells. Subcellular localization assay indicated that SF1 overexpression (both WT and mutated forms) resulted in a redistribution of endogenous SRSF2 to cytoplasm granules and, in some cases, nuclear granules, compared to non-transfected HeLa cells. Confocal microscopy confirmed the presence of nuclear granules containing SRSF2 and revealed that cytoplasmic granules follow a distribution pattern: large irregular granules surrounded by smaller circular ones. What initially appears as a single large granule is actually a cluster of multiple irregular granules. To determine whether there is a difference in the number of cells with SRSF2 granules among the overexpressed conditions (SF1 WT vs. mutations), the observed phenotypes were quantified. Three distinct phenotypes were identified across all transfected cells: SRSF2 localization in: i) nuclear granules (NG); ii) cytoplasmic granules (CG) and; iii) normal localization (NS). After quantifying the cells displaying each phenotype in each condition, the relative percentage was calculated. In SF1 overexpressed cells, those expressing SF1W265L, SF1R345C, SF1G372V and SF1T454M mutations exhibited a significantly higher percentage of cells with the CG phenotype compared to cells expressing SF1WT (p < 0.0001). In the subcellular localization assay conducted with the transduced cells, SRSF2 did not localize in granules and was found in nuclear speckles as well as the non-transduced condition. However, cells overexpressing SF1WT exhibited lower SRSF2 fluorescence intensity compared to those with empty MIG (p< 0.001), and both SF1G372V and SF1T454M mutations had a higher SRSF2 fluorescence intensity value compared to SF1WT (p< 0.001). In the splicing assay, HeLa cells overexpressing the SF1G372V mutation showed no significant difference in splicing activity compared to cells overexpressing SF1 WT (p=0.16). In contrast, HeLa cells transduced with the SF1T454M mutation exhibited a significant reduction in the splicing of the reporter gene compared to SF1 WT (p=0.04). These findings suggest that SF1 may regulate SRSF2 expression, and this regulation could be altered by SF1G372V and SF1T454M mutations. These mutations may alter the splicing function of SF1 directly or influence splicing through SRSF2, with potential splicing changes mediated by SRSF2 itself.