Introduction
Osteosarcoma (OS) is a highly malignant tumor that primarily affects the pediatric and adolescent populations. Efforts have largely centered on understanding the mechanisms associated with its metastases [
1,
2], elucidating the molecular processes governing its development, and identifying potential therapeutic targets to reduce relapse and mortality rates [
3‐
5].
In addition to traditional molecular pathological mechanisms, recent oncological studies have placed increased emphasis on alternative splicing (AS) – a cellular process responsible for generating multiple messenger RNA/protein isoforms from a single transcript [
6,
7]. The splicing factor, a pivotal player in AS progression that orchestrates a multitude of gene transcripts and incites abnormal biological function, has been attributed as a potential harbinger of tumorigenesis [
8].
Noteworthy within this context is the Serine/Arginine-Rich Splicing Factors (SRSFs), a group of structurally related proteins defined by an RS domain rich in arginine and serine residues, integral to efficient alternative RNA splicing [
9]. Of particular interest is SRSF1, an exemplary splicing factor known to bind specifically to exonic enhancers and generate splicing variants [
10]. Numerous studies have investigated the role of SRSF1 in tumorigenesis, with a focus on its involvement in transcriptional regulation mediated by lncRNA or microRNA in breast cancer [
11‐
13], liver cancer, and lung cancers [
14‐
16]. However, the association between SRSF1 and OS has received limited attention, and the precise biological functions of SRSF1 in OS remain largely unexplored. Therefore, this study aims to investigate the role of SRSF1 in the pathogenesis, progression, and development of OS, with a particular focus on its mechanisms and gene-splicing capabilities.
In this study, we conducted in vitro experiments to examine the cancer-related functions of SRSF1 in OS. We analyzed RNA sequences obtained from downregulated SRSF1 in human U2OS osteosarcoma cell lines to explore the underlying mechanisms. Our results provide compelling evidence suggesting that SRSF1 may act as an oncogenic factor, promoting the growth and progression of OS. Consequently, we propose that SRSF1 has the potential to serve as a novel and promising therapeutic target for OS.
Methods and materials
Cell culture
The human osteosarcoma cell lines (143B, MG63, HOS, U2OS) and bone marrow mesenchymal stem cells (BMSC) were purchased from ATCC (Manassas, VA, USA). Osteosarcoma cell lines were grown in DMEM (Gibco, NY, USA), and bone marrow mesenchymal stem cells were grown in low glucose-DMEM (Gibco, NY, USA), supplemented with 10% fetal bovine serum (PAN, seratech, Germany) and 100 U/ml penicillin/streptomycin solution (Gibco, NY, USA) at 37℃ in a humidified 5% CO2 atmosphere.
Gene over-express or knockdown
The lentiviral vector used to overexpress SRSF1 was purchased from GENECHEM (Shanghai, China). Stable cancer cells were established after screening with puromycin (Beyotime Biotechnology, China). The small interfering RNAs used to knock down SRSF1 were manufactured by Ribobio (Guangzhou, China). The siRNA sequences are as follows. siSRSF1#1: CGACGGCTATGATTACGAT; siSRSF1#2: GCAGTTCGAAAACTGGATA; siSRSF1#3: GTACGGAAAGAAGATATGA. The second and the third sequence is used to knockdown SRSF1 for RNA-seq experiment. All the transient transfection was finally performed in HOS cells with the help of Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA). Small interfering RNAs were introduced into cells at a final concentration of 50 nM. The cells were harvested 48 h after transfection.
RNA extraction and quantitative real-time PCR
Total RNA was extracted using Steady Pure Universal RNA Extraction Kit (ACCURATE BIOLOGY, China) according to the manufacturer’s instructions. The concentration and quality of the total RNA were assessed with Nanodrop One Spectrophotometer (Thermo Fisher Scientific, USA). According to the mRNA expression, reverse transcription was performed using Prime Script RT master mix (TaKaRa, Japan). Quantitative real-time PCR analysis was performed in triplicate on 7500 Fast Real-Time PCR System (Applied Biosystems, USA) using SYBR Premix Ex Taq (TaKaRa, Japan) and the expression level of GAPDH was used as endogenous control. To detect the efficiency of transient transfection, β-actin was used as endogenous control. Primers for SRSF1 (forward, 5’-TCTACTGACAGCCCCTTGGT-3’, reverse, 5’-ACTTCCAACTATGATTAGCACCCA-3’), GAPDH (forward, 5’-GCACCGTCAAGGCTGAGAAC-3’, reverse, 5’-TGGTGAAGACGCCAGTGGA-5’) and β-actin (forward, 5’-TGGCACCCAGCACAATGAA-3’, reverse, 5’-CTAAGTCATAGTCCGCCTAGAAGCA-3’). Primers for SRRM2 NM_016333.4 (forward, 5’-TTAAGCCAGGAGCCAGTGAAC-3’, reverse, 5’-CTCGGGAGACTTAGGTGGTGAA-3’) and SRRM2 XM_054379978.1 (forward, 5’- CCGTTCAACTTCTGCTGACTCT-3’, reverse, 5’-CGTGTCTTCCGAATGGTCTGT-3’). Primers for DMKN NM_001035516.4 (forward, 5’-TCTGCTCTGCTCCTGCTCCT-3’, reverse, 5’-GTAGTTCTGATCGTCTCTGCCTGC-3’), and DMKN NM_001190347.2 (forward, 5’-AATCTGGGATTCAGGGGCAAG-3’, reverse, 5’-TAGAGGAGGGCTCGGGTG-3’). Primers for SCAT1 NR_110848.1 (forward, 5’-CCTGGAATAGAAGATGCCTTGG-3’, reverse, 5’-TGCCTAACTTCCTCCTCTAACAA-3’), and SCAT1 NR_110849.1 (forward, 5’-GACTTCTCGTGGGCGTGAGTTT-3’, reverse, 5’-GACCTCGAATGCAACGTCTTCAGAT-3’). Results were analyzed using the 2–ΔΔct calculation method.
Western blotting
Cells were lysed by using RIPA lysis buffer (CWBIO, China). After adding phosphatase inhibitor (100×) and Protease Inhibitor Cocktail (100×), the cell lysate underwent ultrasonication. The protein extracts were separated using 12% SDS gel electrophoresis and were then transferred to PVDF Western Blotting Membranes (Roche Diagnostics Gmbh Mannheim, Germany). After blocking with 5% skimmed milk, the membranes were incubated with primary antibodies of SRSF1 (1:1000, Proteintech), β-actin (1:1000, Proteintech), and GAPDH (1:1000, Cell Signaling), and cultured using secondary antibodies. Protein blots were cut prior to hybridisation with antibodies during blotting. Finally, the experimental results were visualized by using SAGECREATION and analyzed by Image J.
Transwell assay
A Transwell assay was carried out to detect the migration and invasion capacity of HOS and U2OS osteosarcoma cell lines. The assays were performed in 8 μm pore size Transwell chambers (Corning, USA). A total of 100 µl transfected HOS (1.8 × 105/ml) or U2OS (1.8 × 105/ml) cell suspensions without FBS were seeded in chambers, which invasion group was also supplemented with serum-free medium and had an insert coated with Matrigel (Corning, USA). The bottom chamber was filled with DMEM containing 20% FBS. 18 h later, cells were immobilized with Paraformaldehyde and stained with 0.1% crystal violet (Beyotime Biotechnology, China). The experimental results were visualized with a bright-field microscope (Leica DMI4000B, Germany) and analyzed by Image J. And cells invaded in five randomly chosen fields were counted.
Flow cytometry analysis (FCA)
Flow cytometry was used to detect apoptosis and cell cycle which were performed following the manufacturer’s protocol. In brief, cells were washed three times with cold PBS and then resuspended in 500 µl of 1× Binding Buffer, then 5 µl of APC Annexin V and 5 µl of propidium iodide (PI) were added to stain for 15 min at room temperature in the dark. Cells were analyzed by flow cytometry (Cyto FLEX S, Beckman, Germany). For cell cycle analysis, the cell suspension was first fixed with 70% cold ethanol for 3 h and then resuspended in 500 µl of 1× Binding Buffer. After adding 5 µl of propidium iodide (PI) and 15 min room temperature incubation, cells were analyzed by flow cytometry (Cyto FLEX 2, Beckman, Germany). All results were analyzed with FlowJo software (Tree Star).
RNA-Seq
The sequencing data were analyzed with the assistance of Gene Chem. (Shanghai, China). Differentially expressed genes were analyzed by using Deseq2 software with P < 0.05 and |log2(fold change) | >1. The details of RNA-seq were declared in supplementary files.
Statistical analysis
To compare the statistical significance between groups, a two-tailed Student’s t-test, and a one-way ANOVA test were used. It was considered significant when P < 0.05 of each difference. All statistical data were displayed as means ± standard deviation (SD) and analyzed for statistical significance with GraphPad Prism 8 (GraphPad Software, USA).
Discussion
In this study, we investigated the expression and functional role of SRSF1 in osteosarcoma (OS). Consistent with previous research on other types of tumors [
23,
24], we found that SRSF1 expression was significantly elevated in OS tissues and cell lines. These findings suggest that SRSF1 may contribute to the development of OS.
To assess the phenotypic effects of SRSF1, we conducted a series of in vitro experiments using human OS cell lines. We observed that downregulation of SRSF1 resulted in decreased proliferation, migration, and invasion, as well as increased cell apoptosis. Conversely, overexpression of SRSF1 enhanced cell growth, migration, invasion, and anti-apoptosis. These findings indicate that SRSF1 plays a crucial role in promoting the progression of OS. In addition to SRSF1, other splicing factors have been investigated in relation to OS development. For instance, SFPQ has been shown to regulate alternative splicing (AS) of cell cycle-related genes, promoting OS progression [
25]. SRSF3, another member of the SRSF family, has been found to enhance cell viability, migration, and invasion in OS cell lines [
26]. On the other hand, RBM10 has been identified as a tumor suppressor in various types of tumors, including OS, inhibiting proliferation and promoting apoptosis [
27]. As for SRSF1, it has been primarily studied for promoting tumor progression via alternatively splicing RNA as a splicing factor [
12,
23] or for its role in binding to ncRNA to regulate the transcription of other molecules. In the present study, our findings demonstrate that SRSF1 exerts its oncogenic effects in OS through loss- and gain-of-function experiments, promoting proliferation, migration, invasion, and anti-apoptosis. These results align with previous research on the biological functions of SRSF1 in lung and breast cancer. It has been shown that SRSF1 is associated with developmental disorders in lung cancer [
28], and influences patients’ radioresistance [
29]. SRSF1 also exerts oncogenic roles in breast cancer partially by regulating apoptosis and cell proliferation [
11,
30], and is correlated with tumor grade and poor prognosis [
12]. In this study, we reported, for the first time, that SRSF1 functions as a potential oncogene in OS development. To further understand the mechanism of its carcinogenesis, we carried through transcriptome sequencing (RNA-seq).
RNA-seq is commonly employed to evaluate the differential expression of molecules in diseases, serving as a valuable tool for identifying AS [
31]. In this study, we leveraged RNA-seq and identified 1701 upregulated genes and 1317 down-regulated genes in SRSF1-knockdown U2OS cell lines. Through Gene Ontology (GO) analyses, we observed potential associations between SRSF1 and the localization and function of the cytosolic ribosome, protein targeting processes, as well as the composition of the extracellular matrix (ECM) and proteinaceous extracellular matrix. These matrices have emerged as promising avenues for cancer diagnosis and therapeutic targets [
32]. Previous investigations have explored their involvement in breast cancer progression and metastasis [
33], association with the poor prognosis and resistance in non-small cell lung carcinoma [
34], and facilitation of invasion in gastric cancer [
35]. ECM has been implicated in OS development by promoting abnormal bone and vessel activity [
36,
37]. This underscores the significance of SRSF1 in the progression of OS. Based on the KEGG bubble charts, it can be inferred that SRSF1 mainly participates in the PI3K-AKT signaling, Rap1 signaling, and Wnt signaling pathways, focal adhesion, and microRNAs in the development of OS. The Gene Set Enrichment Analysis (GSEA) aligns with the KEGG findings and reveals several other biological signaling pathways, such as PI3K-AKT-related pathways, Wnt, NOTCH, HIPPO, and other metastasis-related pathways, which are closely associated with SRSF1 levels. Intriguingly, the relationship between SRSF1 and some pathways has been verified in previous studies on various tumors. For instance, in hepatocellular carcinoma (HCC), SRSF1 can enhance KLF6 alternative splicing through the phosphoinositide 3-kinase (PI3K)/Akt signaling pathway, generating three splice variants that expedite tumor progression and metastasis [
38]. Additionally, SRSF1 stimulates β-catenin accumulation by recruiting β-catenin mRNA and facilitating its translation in an mTOR-dependent manner, contributing to tumorigenesis [
39]. Furthermore, SRSF1 activates alternative splicing of Numb, an inhibitor of the NOTCH signaling pathway, thereby augmenting NOTCH signaling and promoting tumor growth and progression [
40]. Moreover, some pathways identified in our study have been validated as contributors to the malignant behavior of OS. For instance, the PI3K-AKT signaling pathway, induced by various genes, has been demonstrated to augment chemoresistance in OS [
41] and expedite the progression of OS through mechanisms such as apoptosis suppression and promotion of cell proliferation, migration, and invasion [
42]. The Wnt signaling pathway is also recognized for fostering OS malignancy [
43], fibroblastic traits [
44], distant metastases, and poor survival rates [
45]. The HIPPO signaling pathway is also implicated in the initiation and advancement of OS by stimulating cell proliferation, migration, and invasion in osteosarcoma [
46,
47]. Taken together, these findings suggest SRSF1 may contribute to the progression of OS via biological signaling pathways. Nonetheless, further investigations are required to elucidate the specific pathways through which SRSF1 exerts its effects.
In the context of SRSF1’s AS profile, we observed 766 AS events, with skipped exon (SE) events being dominant, constituting 60% of all events, consistent with previous research indicating SE events as the most prevalent during alternative splicing [
48]. Mutually Exclusive Exons (MXE) events followed, comprising 14% of all events. Upon analysis via GO and KEGG, we inferred that SE-mediated genes through SRSF1 might be implicated in preserving mRNA stability, facilitating RNA degradation, with MXE molecules also seemingly engaged in mRNA stability and RNA transport. Past studies suggest that SRSF1 could activate MAPK signaling, partially due to the upregulation of interleukin 1 receptor type 1 (IL1R1) through alternative-splicing-regulated mRNA stability to trigger pancreatic ductal adenocarcinoma (PDAC) [
23]. SRSF1 has been observed to promote nonsense-mediated mRNA decay (NMD) by recruiting UPF1, suggesting its regulatory role in gene expression and genetic diseases [
49]. Additionally, assessments of Supplementary Tables
3 and Supplementary Table
4 reveal an array of molecules undergoing SE and MXE, some of which have been studied in the context of carcinogenesis in other tumor types. For example, RNMT (RNA guanine-7 methyltransferase) demonstrated an association with CDK1-cyclin B1 to sync mRNA G1 phase transcription and impact mRNA surveillance [
50] and it has also been found to potentially correlate with immune cell infiltration in breast cancer (BC) [
51]. FIP1L1, an MXE molecule managed by SRSF1, was reported as crucial in governing the glycolipid metabolism of GBM cells [
52]. The skipping FIP1L1 (exon 13) modulated by CDYL2a was found to promote cell proliferation in breast cancer [
53]. And FANCI, a DNA repair protein, was found to bind with an apoptotic effector, thereby regulating DNA repair and apoptosis [
54]. PRP3 knockdown causes skipped exon 9 of FANCI and switches the FANCI splicing isoform from FANCI-12 to FANCI‐13 then resulting in delayed DNA damage repair and cell cycle G2/M arrest [
55]. In this study, we confirmed that SRSF1 downregulation alters the abundance of transcripts of SRRM2, DMKN, and SCAT1, indicating a potential influence on their alternative splicing. In previous studies, changes of SRRM2 variants were reported to be closely related with Parkinson’s disease [
56,
57]. Although the role of SRRM2 alternative splicing has yet been uncovered in cancers, its aberrant expression and germline mutation was found to contribute to thyroid cancer progression and recurrence [
58,
59]. The DMKN gene, which harbors five splice variants (α, β, γ, δ, and ε), has been previously studied to suggest that suppressing DMKN-β/γ reduces the invasiveness and migratory capabilities of pancreatic cancer cells, potentially impacting the epithelial-mesenchymal transformation (EMT). Furthermore, DMKN-α, encoded by the variant 1 transcript (NM_001035516.4), has been identified as a potential pancreatic cancer oncogene [
60,
61]. Variant 1 transcript of DMKN changes were also detected in SRSF1 knockdown cells. Besides, although alternative splicing of SCAT1 has yet been reported, its expression type has been found to be associated with cancer prognosis and immunotherapy response [
62,
63]. These findings emphasize the importance of these targeted genes and/or their splice variants in tumorigenesis. Altogether, these observations propose the potential SRSF1-driven mechanism in controlling mRNA surveillance, degradation, and subsequent activation of downstream pathways to foster tumorigenesis in OS. Nevertheless, the specific biological roles of AS and its osteosarcoma-related isoforms remain to be defined. Collectively, our findings shine a light on the importance of AS events catalyzed by SRSF1 in safeguarding mRNA stability, directing degradation, and assuming essential functions in RNA transport in OS.
Despite the insightful findings, this study is not without limitations. First, an exploration into the clinical significance of SRSF1 was not conducted. Second, in vivo experiments, essential for validating the in vitro functions of SRSF1, were not executed. Finally, data obtained from transcriptome sequencing, encompassing the function of molecules and biological pathways, necessitates further validation.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.