KIF18A promotes cervical squamous cell carcinoma progression by activating the PI3K/AKT pathway through upregulation of CENPE

2.1 Bioinformatic analysis

We acquired gene expression data and health information from the Cancer Genome Atlas (TCGA) dataset, available at https://portal.gdc.cancer.gov. The information was extracted in level 3 HTSeq fragments per kilobase per million (FPKM) format. The mRNA expression profiles datasets, GSE63514 [13] and GSE9750 [14], were acquired from the database of GEO (http://www.ncbi.nlm.nih.gov/geo) for subsequent analysis. We employed the t -test function from the R package to assess gene significance between the tumor and normal groups. The thresholds for identifying differentially expressed genes (DEGs) were set as | log₂ fold change (FC) | $>$ 1 and a FDR-adjusted p -value 0.05. The DEGs analysis was performed using the DESeq2 and edgeR packages in R. The clusterProfiler package was employed to perform enrichment analyses for Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA). The potential associations between KIF18A and additional proteins were explored by leveraging the STRING database (https://string-db.org/).

We obtained samples of tumors tha embedded in paraffin from 38 patients with CESC and treated at Weifang People’s Hospital between 2022 and 2023. We obtained recently collected healthy tissues from eight individuals with CESC undergoing radical resection at Weifang People’s Hospital. All specimens of the tissue were subjected to verification by two expert pathologists. The Ethics Committee of Weifang People’s Hospital provided the authorization for the investigation. The clinicopathological characteristics of the 38 cases diagnosed with CESC were meticulously documented and recorded in Table 1.

2.3 Cell lines, cell culture, transfection

The cell lines of HcerEpic and three human CESC (C-33A, CaSki, and SiHa) were obtained from the American Type Culture Collection (ATCC). The cells were grown in DMEM (Gibco, USA) with 1% penicillin/streptomycin and 10% fetal bovine serum (FBS, Gibco, USA) combination. The incubation was conducted in an incubator with a fixed temperature of 37^∘C and under a 5% CO2 environment. The KIF18A expression level in cervical cancer cells was assessed using western blot and qRT-PCR. SiHa cells exhibiting high endogenous KIF18A expression were chosen for further investigation. The targeting sequences for KIF18A (5’-GGAGTGATGTATCTAACAATG-3’) and CENPE (5’-GAAGAAATGGAATTGAAATTA-3’) were incorporated into lentiviral vectors, respectively. The lentiviruses for KIF18A overexpression and its negative control were generated by GeneChem (Shanghai, China). The validation of gene suppression or overexpression was verified by employing qRT-PCR and western blot.

2.4 Western blotting

Tissue specimens and cells were utilized for obtaining total protein employing a customized RIPA buffer solution from Kangwei, China. The proteins have been extracted by 12.5% SDS-PAGE from Epizyme in China and then transmitted to PVDF membranes from Millipore in America with a diameter of the pore of 0.45 mM. Following blocking with 5% skimmed milk powder for sixty minutes, the sample was rinsed three times utilizing TBST. The membranes subjected to incubation at 4^∘C overnight with primary antibodies against KIF18A (1:2000, ab72417), CENPE (1:1000, ab133583), p-PI3K (1:500, ab182651), PI3K (1:1000, ab302958), p-AKT (1:1000, ab38449), AKT (1:1000, ab8933), and GADPH (1:1000, ab8245) obtained from Abcam (Shanghai, China). Subsequently, the membranes were subjected to incubation for 2 h with an antibody conjugated to horseradish peroxidase (HRP) at ambient temperature. The detection of target proteins was achieved using an enhanced ECL kit (Bio-rad, USA).

2.5 RNA extraction and qRT-PCR

The TRIzol reagent (Invitrogen, USA) was employed to obtain total RNA from cells and tissues following the directions of the manufacturer. Next, analysis was performed utilizing reverse transcription-PCR and the SYBR Green II kit (Takara, Japan). The relative target gene expression was standardized with respect to the GAPDH expression. The sequences of primer are as follows: KIF18A (forward) 5’-CTAACCGGGCAAAGGACATTA-3’ and (reverse) 5’-ATCTCTGCCTTCTGCTCATTAC-3’; CENPE (forward) 5’-CACCTCATCCAGTTCGCTATT-3’ and (reverse) 5’-CACTCTCTGCTCCTGCATTT-3’; GADPH (forward) 5’-CACCCACTCCTCCACCTTTG-3’ and (reverse) 5’-CCACCACCCTGTTGCTGTAG-3’.

2.6 Cell proliferation assay

Cells were first placed in 96-well plates at a density of 1 × 10⁴ cells per well and cultivated in full medium for the CCK-8 assay. Following 24–72 hours of incubation, each well was supplemented with 10 μ l CCK-8 solution (Boster, USA). Subsequently, the plate underwent three hours of incubation in a cell incubator, followed by determination of the A450 absorption value using enzyme labeling.

2.7 Cell apoptosis assay

Flow cytometry was utilized to perform the assessment of cell apoptosis according to established protocols. Cells underwent culturing at a density of 1 × 10⁶ cells per in 6-well plates well and were subsequently collected. Following two PBS washes, employing a CytoFLEX flow cytometer (Beckman Coulter, USA), the Annexin V-FITC/PI apoptosis detection kit (Beyotime, China) was employed to assess the apoptosis of the cell.

2.8 Scratch wound assay

Cells underwent seeding into 6-well e-plates and overnight culturing. The following day, a 200 μ l pipette tip was utilized to generate a straight scratch, which subsequently underwent PBS washing. Before the cells underwent culturing for an additional one day, a microscope was utilized to observe them, and then they underwent further observation under the microscope.

2.9 Transwell assay

The invasion assay was conducted employing Matri-gel-coated chambers in 24-well Transwell plates with a pore size of 8 μ m. In the superior chamber, a medium free of serum was used to seed a total of 1 × 10⁵ cells, while the inferior chamber was loaded with a medium supplemented with 10% FBS. Following incubation for a complete period of 24 hours, in the superior chamber, the cells were harvested, whereas the invaded cells in the inferior chamber were fixed and subjected to staining employing a solution comprising 0.1% crystal violet.

2.10 Coimmunoprecipitation (Co-IP) assay

The exogenous Co-IP assay was performed by co-transfecting Flag-KIF18A and HA-CENPE into HEK-293T cells. The cells were lysed using an IP buffer containing 20 mM Tris-HCl, 1 mM EDTA, 150 mM NaCl, and 1% NP-40. Additionally, a 1% phosphatase inhibitor cocktail was included in the buffer solution. The lysis process was conducted at 4^∘C for a duration of thirty minutes. Following a centrifugation step at 12,000g and 4^∘C for ten minutes, the supernatant was collected. Subsequently, it was incubated overnight at 4^∘C with magnetic beads (20 μ l) and antibodies (Flag-tag and HA-tag, each measuring approximately 2–3 μ l). The following day, the supernatant was extracted using a magnetic stand. Subsequently, the beads responsible for binding a specific antibody and the target protein were reconstituted in 1 × loading buffer and subjected to denaturation at 95^∘C for 10 minutes. Subsequently, the specimens were subjected to centrifugation at a speed of 12,000 g for 1–2 minutes at a temperature of 4^∘C. Following this step, the resulting supernatant was carefully preserved at - 80^∘C in preparation for immunoblot analysis. The endogenous Co-IP assay was performed in SiHa cells transfected with Flag-KIF18A using either the Flag antibody or IgG, following identical procedures as the exogenous Co-IP assay.

2.11 Statistical analysis

SPSS 25.0 and GraphPad Prism 8.0.1 were employed to conduct analysis for all experimental data. All assays were performed three times, and the outcomes were represented as the mean ± standard deviation (SD). The Student’s t -test was employed for normally distributed data, while the Mann-Whitney test was employed for data with independent distribution. Outcomes were deemed statistically significant if the P value was less than 0.05.

To discover the underlying molecular pathways of cervical cancer, we conducted an analysis of gene expression profiles obtained from three different datasets: TCGA-CESC, GSE63514, and GSE9750. By conducting a threshold of adjusted p -value 0.05 and | log₂ FC | $>$ 1, we determined differentially expressed genes (DEGs). A Venn diagram illustrating the overlapping region of the 180 DEGs identified from TCGA-CESC, GSE63514, and GSE9750 datasets (Fig. 1A). KIF18A expression levels were significantly elevated in many kinds of cancers compared to their respective control tissues in TCGA (Fig. 1B). The volcano plot of the TCGA-CESC dataset revealed that out of the 4139 DEGs, 2564 genes were upregulated, while 1575 genes were downregulated in CESC tissues in contrast to healthy tissues (Fig. 1C). The Volcano plot revealed the identification of 1089 DEGs (596 upregulated and 493 downregulated) from the GSE63514 dataset (Fig. 1D). The analysis using the Volcano plot identified a set of 2006 DEGs (1126 upregulated and 880 downregulated) from the GSE63514 dataset (Fig. 1E). The KIF18A gene was identified as a common DEG and selected for further analysis. Notably, KIF18A expression levels were significantly upregulated in cervical malignancy tissues in contrast to healthy tissues in TCGA, GSE63514, and GSE9750 datasets (Fig. 1F–H). The protein expression of KIF18A in tumor tissues and corresponding normal tissues from 3 pairs of CESC patients was evaluated through western blot analysis (Fig. 1I). Moreover, we performed RNA extraction from 38 CESC tissues and 9 normal tissues, along with 9 paired CESC tissues and their corresponding normal tissues, followed by mRNA expression analysis using qRT-PCR (Fig. 1J, Fig. S1). The results revealed elevated levels of KIF18A protein and mRNA expression in CESC tissues compared to normal tissues.

OPEN IN VIEWER

The KIF18A mRNA and protein expression were subsequently investigated in CESC cell lines and normal cervical epithelial cells. Significantly elevated KIF18A mRNA and protein levels were observed in CESC cells in comparison with normal cervical epithelial cells (Fig. 2A, B). To further investigate the KIF18A impact, the SiHa cell line was transfected with KIF18A overexpressing lentivirus and shKIF18A lentivirus. The levels of protein and mRNA expression were consequently evaluated to validate the overexpression and knockdown (Fig. 2C, D). The KIF18A impact on the growth of SiHa cells was further investigated using CCK-8 assays. The KIF18A group exhibited significantly enhanced cell viability, whereas a noticeable decrease in cell viability was observed in the ShKIF18A group (Fig. 2E). The upregulation of KIF18A expression was observed to significantly mitigate the occurrence of apoptotic cells in SiHa cells, and this impact could be reversed by downregulating the expression of KIF18A (Fig. 2F).

OPEN IN VIEWER

Consequently, we evaluate the KIF18A impact on CESC cells’ capability of migration and invasion. The migratory capacity of SiHa cells in the wound healing experiment was significantly increased by the overexpression of KIF18A, while the knockdown of KIF18A resulted in a notable reduction in SiHa cell migration (Fig. 3A). The Transwell assay exhibited that the KIF18A upregulation significantly augmented the invasive potential of SiHa cells, while inhibition of KIF18A led to a remarkable reduction in cell invasion (Fig. 3B).

OPEN IN VIEWER

The TCGA-CESC dataset was utilized to identify 618 up-regulated and 17 down-regulated genes that exhibited a strong association with KIF18A based on a screening criterion of | log₂FC $|>$ 1.5 (Fig. 4A, B). These genes were subsequently subjected to a comprehensive analysis using GO and KEGG. The KEGG analysis revealed a significant up-regulated gene enrichment primarily in the PI3K/AKT signaling pathway, while the GO analysis demonstrated enrichment of up-regulated genes predominantly in processes associated with organelle fission and nuclear division (Fig. 4C, D). The KEGG analysis identified a notable enrichment of down-regulated genes mainly within the IL-17 signaling pathway, whereas the GO analysis exhibited that these down-regulated genes were primarily enriched in processes related to antimicrobial humoral response and defense response to bacterium (Fig. 4E, F).

OPEN IN VIEWER

A further discovery from gene set enrichment analysis (GSEA) unveiled a significant PI3K/AKT pathway enrichment in cervical cancer samples exhibiting high expression levels of KIF18A (Fig. 5A). According to the results obtained from KEGG and GSEA analyses, we performed western blot analysis in SiHa cells to explore the KIF18A impact on the PI3K/AKT signaling pathway. The western blot analysis findings provided compelling evidence that the upregulation of KIF18A caused a significant rise in p-PI3K and p-AKT levels while exerting no discernible impact on the total levels of PI3K and AKT expression (Fig. 5B). Furthermore, the downregulation of KIF18A resulted in a reduction in the expression levels of p-PI3K and p-AKT (Fig. 5C).

OPEN IN VIEWER

The TCGA database was utilized to identify co-expressed genes associated with KIF18A in CESC, aiming to obtain a comprehensive understanding of its molecular pathway in CESC. The creation of a heatmap was employed to visually represent the top 20 genes that exhibit a high correlation with KIF18A (Fig. 6A). Our outcomes exhibited a significant positive connection between KIF18A and CENPE expression levels in CESC (Fig. 6B). Additionally, the database of STRING was utilized to generate a protein-protein interaction (PPI) network for KIF18A, unveiling a potential interaction between KIF18A and CENPE (Fig. 6C). The CENPE expression levels were significantly upregulated in cervical malignancy tissues in contrast to healthy tissues in TCGA, GSE63514, and GSE9750 datasets (Fig. 6D–F). The protein expression of KIF18A in tumor tissues and corresponding normal tissues from 3 pairs of CESC patients was evaluated through western blot analysis (Fig. 6G). Moreover, we performed RNA extraction from 38 CESC tissues and 9 normal tissues, along with 9 paired CESC tissues and their corresponding normal tissues, followed by mRNA expression analysis using qRT-PCR (Fig. 6H, Fig. S2). The results revealed elevated levels of KIF18A protein and mRNA expression in CESC tissues compared to normal tissues. The lentiviral infection was utilized to induce stable CENPE knockdown in SiHa cells, thereby confirming the effectiveness of knockdown through qRT-PCR analysis (Fig. 6I). As a consequence, we proceeded to confirm the direct interaction between KIF18A and CENPE. The interaction was validated through endogenous Co-IP assays conducted in SiHa cells (Fig. 6J) as well as exogenous Co-IP assays performed in HEK293T cells (Fig. 6K). Furthermore, Western blot analysis indicated that upregulation of KIF18A expression in SiHa cells led to a significant rise in CENPE levels. Conversely, KIF18A expression downregulation resulted in a notable decrease in CENPE expression (Fig. 6L).

OPEN IN VIEWER

The KIF18A-CENPE axis was further investigated in our study to explore its impact on CESC. The Western blot analysis indicated that the CENPE downregulation effectively attenuated the upregulation of p-PI3K and p-AKT levels induced by KIF18A overexpression (Fig. 7A). Reversal of the enhanced cell invasion resulting from KIF18A overexpression was observed upon silencing of CENPE, as evidenced by the Transwell assay (Fig. 7B). The CCK-8 results demonstrated that the enhanced cellular proliferation induced by KIF18A overexpression was effectively counteracted upon silencing of CENPE (Fig. 7C).