LncRNA KASRT Serves as a Potential Treatment Target by Regulating SRSF1-Related KLF6 Alternative Splicing and the P21/CCND1 Pathway in Osteosarcoma: An In Vitro and In Vivo Study

Purpose Long non-coding RNA KLF6 alternative splicing regulating transcript (lnc-KASRT) locates within the intronic region of SRSF1, possessing the potential to regulate KLF6 alternative splicing to promote carcinogenicity. Then, the current in vitro and in vivo study aimed to investigate the effect of lnc-KASRT on regulating tumor malignant behaviors, and the implication of its interaction with KLF6 alternative splicing in osteosarcoma. Methods Lnc-KASRT overexpression or knockdown plasmid was transfected into U-2OS and Saos-2 cells. Then, KLF6-SV1 knockdown plasmid with or without lnc-KASRT overexpression plasmid was transfected into these cells for compensative experiments. In vivo, lnc-KASRT overexpression or knockdown Saos-2 cells were injected in mice for tumor xenograft construction. Results Lnc-KASRT expression was increased in most osteosarcoma cell lines compared to control cell line. Lnc-KASRT overexpression promoted cell viability, mobility, and anti-apoptotic marker expression, while reducing apoptosis rate and pro-apoptotic marker expression; meanwhile, it regulated SRSF1, KLF6 alternative splicing (increased KLF6-splice variant 1 (KLF6-SV1), decreased KLF6-wild type (KLF6-WT)), and followed P21/CCND1 pathway in U-2OS/Saos-2 cells. The lnc-KASRT knockdown exhibited opposite trends. Subsequent compensative experiments disclosed that KLF6-SV1 knockdown attenuated most of the tumor-promoting effects of lnc-KASRT overexpression in U-2OS/Saos-2 cells. In vivo experiments further validated that lnc-KASRT enhanced tumor growth and reduced tumor apoptosis; meanwhile, it also increased tumor KLF6-SV1, MMP-1, and MMP-9 expressions but decreased tumor SRSF1 and KLF6-WT expressions in xenograft mice. Conclusion Lnc-KASRT serves as a potential treatment target via regulating SRSF1-related KLF6 alternative splicing and following P21/CCND1 pathway in osteosarcoma.


INTRODUCTION
Osteosarcoma is one the most vicious cancers that primarily affect children and young adults, with a low overall incidence rate of 3-5 per million in males and 2-4 per million in females each year (1,2). Due to the exasperated features of osteosarcoma, such as high heterogeneity, rapid growth rate, and strong invasive ability, a nonnegligible proportion of patients are diagnosed with distant metastases, and its prognosis is very dismal (3)(4)(5). To improve the survival of patients with osteosarcoma, continuous and sincere efforts have been made to enhance surgical technology, neoadjuvant/adjuvant treatment strategies, novel targeted drugs, and personalized medicine, and these approaches have progressed to some extent (6)(7)(8)(9). However, the prognosis of osteosarcoma patients is still far from satisfactory. Therefore, it is crucial to explore the underlying mechanism of osteosarcoma development and progression to identify potentially useful treatment targets.
Then, our current in vitro and in vivo study aimed to investigate the effect of lnc-KASRT on regulating malignant tumor behaviors and the implication of its interaction with KLF6 alternative splicing in osteosarcoma.

Tumor Xenograft Model Construction
All animal experiments were approved by the Institutional Animal Care and Use Committees (IACUC) of our institution and were performed in accordance with institutional guidelines and ethical standards. A tumor xenograft model was constructed according to methods described previously (25). Male BALB/C nude mice (5-6 weeks of age, 16-18 g) were maintained under specific pathogen-free conditions. Approximately 2 × 10 6 KASRT(+) cells, KASRT(-), cells and NC cells were suspended in 100 ml PBS and then injected subcutaneously into the right side of the posterior flank of nude mice. The mice were divided into the KASRT(+) group, KASRT(-) group, and NC group. Tumor volumes were examined every 7 days using a Vernier caliper (Mitutoyo, Japan). Tumor volumes were calculated using the equation: Volume = a*b 2 /2 (mm 3 ), where a is the largest diameter and b is the perpendicular diameter. Four weeks after cell injection, mice were sacrificed, and tumors were harvested. After weighing, tumors were stored properly for further analysis. For tumors placed in 10% formalin and embedded in paraffin, hematoxylin-eosin (HE) staining was performed for histological analysis, terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay was carried out to detect cell apoptosis, and immunochemistry (IHC) was performed to detect protein expression of SRSF1, KLF-6-SV1, KLF-6-WT, MMP-1, and MMP-9. For tumors stored at -80°C, SRSF1, KLF-6-SV1, KLF-6-WT, MMP-1, and MMP-9 mRNA expression was evaluated by RT-qPCR assays.

RT-qPCR
Total RNA was extracted by TRIzol ™ Reagent (Invitrogen, USA), and then reverse transcription and PCR were performed with the PrimeScript ™ RT reagent Kit (Takara, Japan) and TB Green ™ Fast qPCR Mix (Takara, Japan), respectively. Gene expression was calculated by the 2 -△△Ct method with GAPDH as the internal reference according to a previously reported method (26). Primer sequences are listed in Supplementary Table 1.

Western Blot
Western blotting was performed by standard procedures according to previous studies (22,23). Protein extraction and quantification were performed using RIPA buffer (Sigma, USA), and a bicinchoninic acid kit was used for protein determination (Sigma, USA). Furthermore, NuPAGE ™ 4-20% Tris-Acetate Midi Protein Gels (Thermo, USA), XCell SureLock ™ Mini-Cell (Invitrogen, USA), Pierce ™ ECL Plus Western Blotting Substrate (Invitrogen, USA), X-ray film (Kodak, USA), and Gel Imager (Thermo, USA) were also used. Antibody information is presented in Supplementary Table 2. HE Staining, TUNEL Assay, and IHC HE staining, TUNEL assay, and IHC experiments were performed according to the methods described previously (26,27). The formalin-fixed and paraffin-embedded tumor specimens were sliced into 4 mm sections. Then, a hematoxylin and eosin staining kit (Beyotime, China) was used for HE staining, which was performed in accordance with the kit's standard protocol. The TUNEL assay was performed using a TumorTACS in situ apoptosis kit (R&D, USA) according to the manufacturer's manual. Furthermore, IHC was completed according to the standard operating protocol of The Early Detection Research Network (EDRN), and the antibodies used for the IHC assays are listed in Supplementary Table 2. Images were obtained using an Olympus BX41 microscope (Olympus, Japan).

Statistical Analysis
Bar charts and line charts with error bars were used to display a statistical summary of the mean value and standard deviation (SD).
Comparison between two independent samples was determined by the unpaired t test; multiple comparisons between a control group and other experimental groups were determined by one-way ANOVA followed by Dunnett's test; multiple comparisons between each group were determined by one-way ANOVA followed by Tukey's test. Statistical analysis and graph plotting were performed using GraphPad Prism 7.02 software (GraphPad Software Inc., USA). Significance was defined as a P < 0.05, which is displayed as *P < 0.05; **P < 0.01; ***P < 0.001, and nonsignificant is marked as NS.

Lnc-KASRT Promoted Osteosarcoma Cell Viability and Mobility
Lnc-KASRT was elevated in several osteosarcoma cell lines, such as U-2OS, MG-63, and MNNG/HOS, but was not changed in the Saos-2 cell line compared to the control cell line hFOB1.19 ( Figure 1A). After transfection in both U-2OS and Saos-2 cells, lnc-KASRT was increased in the KASRT (+) group compared to the NC (+) group but was decreased in the KASRT (-) group compared to the NC (-) group, indicating successful transfection (Figures 1B, C). Regarding cell viability, lnc-KASRT overexpression promoted cell proliferation and expression of the antiapoptotic marker BCL-2 and reduced the cell apoptosis rate and expression of the proapoptotic marker C-Caspase3 in both U-2OS (Figures 2A-D) and Saos-2 cells (Figures 2E-H). In terms of cell mobility, lnc-KASRT overexpression enhanced cell migration and invasion in both U-2OS (Figures 3A-D) and Saos-2 cells (Figures 3E-H). Moreover, lnc-KASRT knockdown exhibited the opposite effect on these cell functions (Figures 2A-H) (Figures 3A-H). However, lnc-KASRT had little effect on chemosensitivity in U-2OS cells, except that it had a slight effect on doxorubicin chemosensitivity in U-2OS cells (Supplementary Figures S1A-C); therefore, subsequent exploration of chemotherapy sensitivity was not continued.

DISCUSSION
Two decades ago, the alternative splicing ability of SRSF1 was revealed, and it has been shown to regulate many complicated biological processes comprising some important aspects of mRNA metabolism (such as mRNA splicing, stability, translation, etc.) and other mRNA-independent pathways (such as miRNA processing, protein sumoylation, nucleolar stress response, etc.) (28,29). On account of the above features, SRSF1 is also considered a key carcinogenic factor that functions via multiple mechanisms, such as regulating extrinsic (for instance, Fas, Caspase-8, Caspase-2, c-FLIP) and intrinsic (Apaf-1, Caspase-9, ICAD) factors, genes encoding Bcl-2 proteins, IAPs, and p53 tumor suppressors (29,30). Interestingly, along with the advances in high-throughput sequencing and bioinformatics, a novel kind of RNA, named lncRNA, was recently discovered. LncRNAs have a sequence length of 200 bp or longer and have no protein-coding abilities, and they function as sponges for miRNAs or their parent/target genes (20). Meanwhile, within the SRSF1 intronic region, UCR-419 (named lnc-KASRT) fits the definition of a lncRNA and regulates KLF6 (a common tumor regulator) alternative splicing by sponging the SRSF1 gene. The role of lnc-KASRT and KLF6 alternative splicing in osteosarcoma development and progression is unclear. Therefore, our current study found that lnc-KASRT promoted osteosarcoma cell viability and mobility; moreover, lnc-KASRT regulated KLF6 alternative splicing to induce KLF6-SV1 while repressing KLF6-WT and regulating P21 and CCND1 expression in osteosarcoma. The possible explanations for these effects were as follows: (1) lnc-KASRT could regulate multiple oncogene/anti-oncogene alternative splicing events, including KLF6, to promote osteosarcoma cell malignant behaviors. (2) lnc-KASRT could directly sponge the SRSF1 gene itself to reduce SRSF1 expression, thus reversing KLF6 alternative splicing and reducing KLF6-WT while enhancing KLF6-SV1. (3) lnc-KASRT could regulate P21 and CCND1 upstream or directly regulate P21 and CCND1 via its alternative splicing ability.
KLF6, a common anti-oncogene, represses cell proliferation, migration, evasion, and epithelial-mesenchymal transition (EMT) and facilitates cell apoptosis and chemoradiotherapy sensitivity by regulating multiple pathways, such as the hedgehog, p53 apoptotic, EGFR, and p21/CCND1 pathways (31)(32)(33)(34). In addition, it is fascinating that when the alternative splicing of KLF6 is disrupted or modified, KLF6 encodes KLF6-SV1 instead of KLF6-WT, which makes it act as an oncogene (19). Accumulating studies have reported that KLF6-SV1 promotes tumor development and progression in various ways, such as by regulating the NOXA and TWIST genes and the PI3K/AKT pathway (14)(15)(16)(17). Considering the effect of lnc-KASRT on SRSF1 and KLF6 alternative splicing, we further explored the role of KLF6-SV1 in osteosarcoma malignant behaviors and whether it would affect the function of lnc-KASRT. Then, we found that KLF6-SV1 knockdown inhibited cell proliferation, migration, and invasion, promoted cell apoptosis, and regulated the P21/CCND1 pathway in osteosarcoma, which was in line with previous studies regarding the effect of KLF6-SV1 in other cancers. More importantly, we further discovered that KLF6-SV1 attenuated the effect of lnc-KASRT on regulating osteosarcoma cell functions and the P21/CCND1 pathway. These results indicated that lnc-KASRT indeed regulated osteosarcoma by modifying KLF6 alternative splicing. Furthermore, to validate the above mentioned in vitro findings, we also performed in vivo experiments and found that lnc-KASRT promoted osteosarcoma growth and invasive markers while reducing tumor necrosis and apoptosis. Moreover, lnc-KASRT negatively regulated SRSF1 and KLF6-WT expression and positively modified KLF6-SV1 expression. These results also validated the notion that lnc-KASRT facilitated osteosarcoma progression by regulating KLF6 alternative splicing.
To make the key findings of this study easy to interpret, a graphical figure illustrating the related mechanism was created ( Figure 11). This visual summary shows that lnc-KASRT located on SRSF1, interacts with SRSF1, then regulates alternative splicing of KLF6 to promote KLF6-SV1 coding while inhibiting KLF6-WT coding, and subsequently modifies P21/CCND1 pathway, finally leading to increased growth and invasion of osteosarcoma.
There were some limitations in the current study. First, clinical human samples were not used for lnc-KASRT, SRSF1, KLF6-SV1, or KLF6-WT expression detection due to the limited number of patients. Second, the viability and mobility capacity were measured in this study; however, since our preliminary experiments observed that lnc-KASRT affects chemotherapy sensitivity little in U-2OS cells, we did not further explore the deep engagement of lnc-KASRT in treatment resistance, which could be considered in the future. Third, KLF6 was not only reported to be a tumor regulator but also an immune/inflammation trigger; therefore, the involvement of lnc-KASRT and KLF6 in the tumor immune microenvironment is another hotspot for further research. Fourth, in silico study to assess the application of lnc-KASRT as a target in osteosarcoma treatment was needed in the future.
In conclusion, lnc-KASRT serves as a potential treatment target by regulating SRSF1-related KLF6 alternative splicing and the P21/CCND1 pathway in osteosarcoma.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors. Lnc-KASRT is located on SRSF1, interacts with SRSF1, and then regulates alternative splicing of KLF6 to promote KLF6-SV1 coding while inhibiting KLF6-WT coding, subsequently modifying the P21/CCND1 pathway, finally leading to increased growth and invasion of osteosarcoma.

ETHICS STATEMENT
All animal experiments were approved by the Institutional Animal Care and Use Committees (IACUC) of our institution and were performed in accordance with institutional guideline and ethical standard.

AUTHOR CONTRIBUTIONS
YC and XZ conceived and designed the study. KC, CL, and SH collected and analyzed the data. KC and CL prepared the figures and tables. KC, CL, and SH wrote the manuscript. YC and XZ edited the manuscript. All authors revised the manuscript and read and approved the submitted version.