Kinesin family member 4A (KIF4A) is highly expressed in endometrial cancer (EC)
Kinesins have been found to be dysregulated in a variety of tumors. To explore the expression levels of KIFs in uterine corpus endometrial carcinoma (UCEC) and normal endometrial tissues, TCGA database and Gene Expression Omnibus data were analyzed. We found 11 common upregulated KIFs (KIF1A, KIF2C, KIF4A, KIF11, KIF14, KIF15, KIF18A, KIF18B, KIF20A, KIF23, and KIFC1) and one common downregulated KIF (KIF26A) in TCGA-UCEC and GSE17025 databases (Fig. S1A, B). By detecting the mRNA level of kinesin in EC tissue samples, we found that KIF4A increased most significantly in EC (fold change = 3.64, P < 0.01) (Fig. 1A). Furthermore, in TCGA and GSE17025 databases, the mRNA expression of KIF4A was significantly upregulated in EC tissues (Fig. 1B, C). We also selected 23 paired samples from TCGA database and performed a paired t-test on the expression level of KIF4A, which was consistent with the overall results (Fig. 1D). Interestingly, we observed that KIF4A was elevated in all cancer types, compared to that in normal tissues (Fig. 1E). Regarding protein levels, western blot results showed that KIF4A was increased in human EC compared to adjacent normal tissues. In addition, immunohistochemical staining for KIF4A was performed in paraffin sections obtained from 20 normal and 35 tumor tissue samples, which suggested that KIF4A was highly expressed in EC tissues (Fig. 1G).
KIF4A acts as an indicator of unfavorable outcomes in patients with EC
We further explored the prognostic value of KIF4A in patients with UCEC by combining clinical data from the TCGA database. The patients were divided into two groups (high vs. low) based on the median expression of KIF4A across all samples. Overall survival (OS), disease-free interval (DFI), disease-specific survival (DSS), and progression-free interval (PFI) were estimated using the Kaplan–Meier method, and the results indicated that the high expression of KIF4A led to statistically significant inferior outcomes compared with that of the low KIF4A expression group (OS: HR, 1.750, 95% confidence interval (CI)= 1.153–2.658; PFI: HR, 2.026 ,95% CI = 1.417–2.897; DSS: HR, 1.978 ,95% CI = 1.183–3.305; DFI: HR, 1.925 ,95% CI = 1.137–3.259) (Fig. 2A–D). Next, we analyzed the correlation between KIF4A expression and the clinicopathological features of the patients (Table 1). The expression level of KIF4A was highly correlated with the pathological type, clinical stage, histologic grade, age, and vital status in patients with EC (Fig. 2E–I). Receiver operating characteristic curves were used to analyze the diagnostic value of KIF4A in TCGA-UCEC and GSE17025 databases. As shown in Fig. 2J–L, KIF4A had a high area under the curve value in patients with EC, indicating that KIF4A could effectively distinguish EC tissues from normal tissues. Collectively, these results indicate that KIF4A is upregulated and predicts poor prognosis in EC.
KIF4A promotes EC cell proliferation
We next determined KIF4A protein expression levels in normal human endometrial cells and several human EC cell lines by western blotting and quantitative polymerase chain reaction. Subsequent analyses were restricted to the KLE and Ishikawa cell lines because of their heightened KIF4A expression (Fig. 3A, B). Therefore, to elucidate the function of KIF4A in cell proliferation, we designed three independent shRNAs to knock down KIF4A expression in KLE and Ishikawa cells. KIF4A silencing was verified by quantitative real-time polymerase chain reaction (qRT-PCR) and western blotting (Fig. 3C, D). Colony formation, Cell Counting Kit-8 (CCK-8), and 5-ethynyl-2′-deoxyuridine (EdU) assays were used to measure the effect of KLE and Ishikawa cell proliferation and growth, which indicated that the cell growth and colony formation potential were significantly reduced in EC cells with stable expression of KIF4A shRNA (Fig. 3E–G).
Knockdown of KIF4A leads to cell cycle arrest and apoptosis
To further clarify the role of KIF4A in cell proliferation regulation, we performed gene set enrichment analysis (GSEA) of the TCGA-UCEC database, which revealed that the top signatures in the high KIF4A group were strongly associated with cell cycle andapoptosis (Fig. S2A–C). Thus, we hypothesized that KIF4A may be involved in regulating cell cycle progression. Flow cytometry was performed to evaluate its effect on the cell cycle status. As hypothesized, KIF4A knockdown increased the proportion of cells in the G2/M phase and decreased that in the S phase (Fig. 4A). Western blot results verified that the protein levels of the G2/M phase marker, cyclin B1, and CDK1 also decreased in EC cells with KIF4A knockdown (Fig. 4B). Next, the effect of KIF4A on EC cell apoptosis was measured by flow cytometry apoptosis assay, which demonstrated that KIF4A knockdown significantly increased the apoptosis rate (Fig. 4C). Similarly, KIF4A knockdown upregulated the expression of Bax, cleaved caspase 3, and decreased the expression of Bcl-2 (Fig. 4D). These results suggest that KIF4A knockdown induces G2/M phase arrest and promotes apoptosis of EC cells.
KIF4A knockdown activates DNA damage response signaling
To explore the downstream functional roles of KIF4A, RNA sequencing was performed in KLE cells stably knocked down KIF4A (sh-KIF4A vs. sh-Control). A total of 1621 differentially expressed genes (DEGs) were identified in cells stably knocked down KIF4A compared with sh-Control, with 1286 genes upregulated and 335 downregulated(Fig. 5A, B). To further analyze the DEGs, pathway enrichment analysis was performed, and the results revealed that mitotic G2-G2/M phases, cell cycle checkpoints, apoptosis, nucleotide excision repair, and DNA double-strand break repair pathways were activated simultaneously (Fig. 5C). Subsequently, Gene Ontology enrichment analysis indicated that these genes were predominantly enriched in cell cycle, DNA repair, apoptotic signaling pathway, G2/M transition of mitotic cell cycle, double-strand break repair, and ubiquitin (Ub)-dependent protein catabolic process (Fig. 5D). Furthermore, GSEA indicated that KIF4A was also highly associated with homology-directed repair, chromosome maintenance, DNA double-strand break response, G2/M DNA damage checkpoint, and Hub-specific processing protease pathway in the TCGA-UCEC database (Fig. S2D–I). These results suggest that KIF4A may be involved in DDR signaling.
DDR is composed of a coordinated network of signaling pathways that sense DNA damage and induce cellular responses through a set of DNA repair mechanisms. Double-strand breaks (DSBs) are the most deleterious DNA damage that leads to chromosomal aberrations, genomic instability, or cell death. Phosphorylation of H2AX (γH2AX) plays a key role in DDR as γH2AX is necessary for the DNA repair process, and γH2AX activates checkpoint proteins that arrest cell cycle progression[25]. Histone H2AX is a substrate of several phosphoinositide 3-kinase-related protein kinases, such as ataxia-telangiectasia mutated (ATM), Rad3-related kinase, and DNA-dependent protein kinase[26]. DDR signaling was assessed by examining the phosphorylation status of key DDR signaling molecules, including histone H2AX (γH2AX), ATM, and CHK2[27]. During mitosis of eukaryotic cells, the separation of sister chromosomes is mainly mediated by kinesins, which bind to chromosomes, move along microtubules, and exert tension and thrust on centrosomes[28]. Therefore, kinesins are important for the maintenance of genomic stability. Previous studies have also shown that KIF2C depletion or inhibition leads to the accumulation of endogenous DNA damage[29]. KIF18B and KIFC1 have been reported to promote 53BP1-mediated DNA double-strand break repair[30] .
Therefore, to further elucidate the pathway that mediated cell apoptosis, we examined
the effect of KIF4A knockdown on phosphorylation of H2AX, ATM, and CHK2. Western blotting showed that KIF4A knockdown EC cells had increased DDR pathway activity, as evidenced by elevated pATM, pCHK2, and γH2AX levels. Furthermore, immunofluorescence staining was performed using an anti-γH2AX antibody, which suggested that EC cells with stable KIF4A knockdown could form more numerous DNA-damage-induced nuclear foci. Collectively, these results indicate that KIF4A knockdown causes DNA damage accumulation and upregulates cellular DDR-related gene expression.
KIF4A regulates the protein stability of TPX2
To identify the association between KIF4A and DDR signaling pathways in EC cells, we utilized a system‐level approach to understand functional protein interactions (http://string-db.org/), and the top 10 KIF4A-interacting proteins were identified (Fig. 6A). We then set out to confirm whether an interaction exists between KIF4A and these 10 proteins. We used western blotting to examine the protein levels of these proteins in Ishikawa and KLE cells with KIF4A knockdown. Notably, CDK1 and TPX2 expressions were significantly downregulated in EC cells after KIF4A knockdown compared to that in control cells (Fig. 6B, Fig. S3A). Moreover, the qRT-PCR results showed that KIF4A knockdown repressed CDK1 mRNA levels but had no effect on TPX2 mRNA levels (Fig. 6C, Fig. S3B). The inconsistency between the protein and mRNA levels of TPX2 suggests that KIF4A may regulate TPX2 levels by regulating the stability of TPX2 protein.
To further verify whether KIF4A interacts with TPX2, coimmunoprecipitation was performed, and the results revealed that endogenous KIF4A could bind to TPX2 in EC cells (Fig. 6D, E). Subsequently, KLE and Ishikawa cells with stable knockdown of KIF4A were treated with cycloheximide (CHX) to inhibit protein synthesis, and TPX2 protein turnover was analyzed over time. Compared with that in the control cells, the TPX2 half-life was considerably decreased in EC cells with stable knockdown of KIF4A, which were treated with CHX (Fig. 6F). The Ub-proteasome and autophagy-lysosome pathways are the two main routes of protein degradation in eukaryotes[31]. To determine which pathway played a major role in this process, a proteasome inhibitor (MG132) and a lysosome inhibitor (chloroquine) were administered to KIF4A knockdown cells. Strikingly, only MG132 completely restored the TPX2 protein level in KIF4A knockdown EC cells compared to that in the untreated control (Fig. 6G).
Our data suggest that knockdown of KIF4A represses TPX2 protein stability mainly by promoting protein degradation mediated by the Ub/proteasomal pathway. Next, we analyzed the effect of KIF4A knockdown on the ubiquitination of endogenous TPX2 through both immunoprecipitation and western blotting, which showed that the level of ubiquitinated TPX2 was significantly higher in EC cells with the stable knockout of KIF4A compared to control cells (Fig. 6H). In general, the above results indicate that KIF4A regulates TPX2 protein stability by affecting TPX2 ubiquitination levels.
TPX2 is required for KIF4A-mediated tumor progression
TPX2 is a multifunctional protein that includes one or more potential microtubule-binding sites that play a vital role in spindle assembly[32]. Other recent studies also reported that TPX2/Aurora A could protect DNA forks during replication stress[33].
To further verify whether KIF4A induced EC progression by regulating the levels of TPX2, rescue experiments were performed. We co-transfected sh-KIF4A and pcDNA-TPX2 into KLE and Ishikawa cells. Next, we examined the effect of TPX2 overexpression on KIF4A knockdown-induced suppression of proliferation in vitro. The CCK-8 assay indicated that TPX2 overexpression could reverse the KIF4A knockdown-induced inhibition of proliferation (Fig. 7A, B). The EdU assay showed that TPX2 overexpression could rescue the decrease in EdU-positive cells induced by KIF4A knockdown (Fig. 7C). Moreover, we examined the influence of TPX2 overexpression on KIF4A knockdown-induced cell cycle arrest and apoptosis, which indicated that TPX2 overexpression impairs the KIF4A knockdown-mediated increase in EC cell cycle arrest and apoptosis (Fig. S3C). Furthermore, immunofluorescence staining of γH2AX also revealed that TPX2 overexpression markedly reduced the number of γH2AX foci (Fig. 7D). Meanwhile, a western blot was performed to probe the effect of TPX2 overexpression on the DDR signaling pathway, which suggested that TPX2 overexpression could suppress the expression of pATM, pCHK2, and γH2AX induced by KIF4A knockdown (Fig. 7E). Based on the above observations, we concluded that KIF4A mediated tumor promotion through TPX2-dependent signaling.
A subcutaneous xenograft tumor model is used to confirm the role of KIF4A in EC
Based on the observations described above, we examined the effects of KIF4A on EC cell proliferation in vivo by stably transfecting KLE cells with sh-KIF4A or sh-NC and subcutaneously injecting into nude mice (n = 5 per group). Tumor volume was measured every 4days after injection. After 32 days, the mice were euthanized, and whole tumors were excised. The results demonstrated that KIF4A knockdown significantly reduced both the weight and volume of tumors in vivo (Fig. 8A–C). Immunohistochemical staining was performed to evaluate the expressions of KIF4A and TPX2 in xenograft tumors, and the results were consistent with the in vitro findings. The decrease in KIF4A expression was accompanied by decreased levels of TPX2 (Fig. 8D). The expression of ki-67, a proliferation marker, was also decreased in sh-KIF4A EC tumors. Collectively, these results suggest that KIF4A functions as a tumor promoter in EC and that KIF4A promotes EC cell progression via TPX2.