The role of CENPO expression in the progression of non-small cell lung cancer

Background: The role of Centromere protein O (CENPO) in the development of non-small cell lung cancer (NSCLC) is unknown. This study aimed to investigate the potential roleof CENPO in the development of NSCLC Methods: The expression level of CENPO was investigated in both tissues and cell-lines. Celigo cell counting assay, wound-healing assay, Transwell assay and Flow cytometry was used to explore the effect of CENPO on proliferation, migration, invasion, apoptosis and cell cycle of NSCLC. The potential mechanism of CENPO was explored by Human Apoptosis Antibody, also, western blot was conducted to detect the expression of PI3K/Akt/mTOR pathway and cell-cycle related protein (mTOR, P-mTOR, CDK1, CDK6 and PIK3CA). Besides, the effect of CENPO on the growth of NSCLC solid tumors was demonstrated in vivo. Results: Our study suggested CENPO gene overexpression in NSCLC. Reduced CENPO expression substantially decreased the proliferation, migration and invasion ability, and promoting apoptosis and induces cell cycle arrest of NSCLC cell-lines. Preliminary mechanism research suggested reduced CENPO could active apoptosis pathway, suppressing PI3K/AKT/mTOR pathway and down-regulation cell-cycle related proteins. Conclusion: was up-regulated and played important the of


Introduction
Themorbidity and mortalityo ung cancer was the highest worldwide, with 2.1 million new lung cancercases and 1.8 million deaths in 2018, representing close to 1 in 5 (18.4%) cancer deaths based onthe GLOBOCAN 2018. (1)In China, lung cancer was also still a fatalthreatto people's heath, with the incidence of 774,323 and the mortalityof690,567 in 2018. (2) Non-small cell lung cancers (NSCLCs) account for approximately 85% lung cancer cases, which could be divided into 2 major type -lung adenocarcinoma (LUAD) and lung squamous cell carcinoma(LUSC).(3) Nowadays it is accepted that LUAD is the dominant subtype, especially in Asian women (more than 70% in Japanese females).
Although many advances have been made in the early detection and treatment of lung cancer, the prognosis remains poor, with an overall 5-year survival rateranging from 15% to 20%.Recently, manytargetable gene alterations haven been identi ed to bene t from targeted therapy and immunotherapy, however,not all lung cancer have targetable gene alterations, and many patients develop resistance to targeted therapy. The heterogeneity was a majorproblem and achallenge for lung cancer treatment. The reason for this heterogeneity might be the difference in molecular features and biological behaviorsamong lung cancers. With the development in molecular biology, for example, the use of nextgeneration sequencing, we couldmake it possible in understanding the distinct molecular feature in lung cancer.
Correct transmission of genetic material during the chromosome segregation in cell division iscritical for the fatal of cells.Errors inchromosome segregationmayresult in some genetic disease, alsoneoplasia.The formation of kinetochore isessential in the process of chromosome segregation. (4) Recent studies have demonstrated that kinetochore is a macromolecular complexconsisted of many proteins. The kinetochore protein complex can be roughly divided into constitutive centromere-associated network (CCAN) and KMN (KNL1-Mis12 complex-Ndc80 complex) network. ( In our current study, we focused on centromere protein O (CENPO), which formed CENP-O/P/Q/U complex. Previous study showed CENP-O/P/Q/R/U-de cient chickenDT40 cells are viable, but show defects during the process of recovery from spindle damage (7). Depletion of CENPO proteins by siRNAfrequently induced aneuploidy in RKO cells. Accumulation of aneuploid chromosomes would leads to the development of cancer.(8) The researches about the role of CENPO and the potential mechanism in cancer were limited. Only Cao Y et. al reported that CENPO wasupregulated ingastric cancer and was associated with poor survival. Reduced CENPO contributed to cell growth inhibition and apoptosis induction. (9)Further studies were needed to validate the role of CENPO in other cancers and explore its potential role.
In the current study, we comprehensively investigated the role of CENPO expression in the progression ofNSCLC.Firstly, we examined the expression level of CENPO in NSCLC tissues and cells. Secondly, we explored the role of CENPO in proliferation, migration, invasion, apoptosis and cell cycle on NSCLC by both using cell-line and orthotropic xenograft models. Finally the preliminarily mechanism of CENPO in the tumorigenesis of NSCLC was investigated.

Western blot
The protein were extracted from cells by 1×Lysis buffer, then the concentration was measured by BCA Protein Assay Kit (Cat 23225, Hyclone-Pierce).10% SDS-PAGE was used to separate protein and then the proteins were transferred to PVDF membrane. The membrane was blocked with1×TBST solution with 5% skimmed milk for 1 h at room temperature and then incubated at 4˚C overnight with primary antibodies. After that, the membrane was incubated with secondary antibodies at room temperature for 2 h. At last, immobilon western Chemiluminescent HRP Substrote was used to visualize the protein expression level.

Celigo cell counting assay
The cells were seeded in wells of 96-well plate (1000 cells/well 3 replicates for each group) at 37°C in an atmosphere of 5% CO 2 . The Celigo Imaging Cytometer (Nexcelom, Lawrence, MA, USA) was used to calculate the living cells by measuring green uorescence in each plate. The measurementbeganafter 24 h incubation and daily measured for 5 days so that the cell growth curves were drawn.

Wound-healing assay
The cells were seeded in wells of 96-well plate at the densityof 50,000 cell/well. After the 24h, cells were scratched with pipette tips, and the ablated cells were washed away with phosphate-buffered saline (PBS). According to the degree of wound-healing, the appropriate timewas chosen to analyze the woundhealingarea by cellomics (ArrayScan VT1, Thermo, USA). The mobility of the cells was determined by measuring the wound-healingarea.

Transwell assay
Cell invasion ability was determined by Transwell assay, a Matrigel-coated modi ed Boyden chamber with a polycarbonate lter with a pore size of 8 μM. 100uL cell suspensions (100,000~200,000 cell) with serum-free medium was added in the upper chamber of the transwell coated with 50 mg/ml of Matrigel solution. High-serummedium (30% FBS+1640) was then added to the lower chamber. After incubation for 24 h, the cells were xed with a 4% formaldehyde solution for 30 mins and stained with giemsa solution. Five random elds in each chamber and photographed with a microscope at 100× magni cation. ImagePro +6.0 was used to calculate the average migrated cells to determine the invasion capacity.

Flow cytometry (FCM)
The impact of CENPO expression on cell cycle and apoptosis was evaluated by ow cytometry (FCM) assay. The cells were seeded in wells of 6-well plate at 37°C in an atmosphere of 5% CO 2 . When the cells grew to 80% con uence, they were collected by centrifugation at 1500 rpm for 5 min at 4°C and washed three times by cooled PBS solution. The cells were stained with propidium iodide (PI) (Sigma P4170) or Annexin V (eBioscience) to determine cell cycle distribution and cellular apoptosis by ow cytometry using a Guava easy Cyte HT ow cytometer (Millipore).

Human Apoptosis Antibody Array analysis
Human Apoptosis Antibody Array (Abcam, Cambridge, MA, USA) was used to detect the effect of CENPO knockdown on the apoptosis pathway-related protein expression. NCI-H1299 cells infected with shCENPO were cleaved with 2 × cell lysis buffer. The membranes were incubated with 1.2 mL cell lysate overnight at 4°C, and then the liquid was discarded. Membranes were washed with 2 mL Wash Buffer I 3 times, and 2mL Wash Buffer II 2 times, each time for 5min. 1mL 1 × Bolin-conjugated Anti-Cytokins were added and incubated with membranes overnight at 4°C. Wash Buffer I and Wash Buffer II were used to wash the membranes according to the above steps. The equal volume detection buffer C and detection buffer D were mixed evenly and added to the membranes for incubation for 2 min at room temperature. The signals were detected by chemiluminescence imaging system. And unpaired t-test was used for statistical analysis.
Nude mouse tumor formation model Four-week old female BALB/c nude mice (Charles River Laboratories, Beijing, China) were divided into shCtrl (infected with shRNA control) and shCENPO (infected with CENPOsiRNA), 10 mice in each group.2 × 10 6 cells were injected intoeach nude mice. Tumor sizes were measured every 5 days using a Vernier caliper. Tumor volume was calculated using the following formula: volume = 0.5 * width 2 * Length. After 39 days, the mice were sacri ced andwere placed under the in vivo imaging system (LB983, Berthold Technologies) toobserve the uorescence. The tumors were removed from the mice, and thenthe weight of the tumors was measured. Tumor tissues were frozen in liquid nitrogen and stored at -80°C for further analysis.

Statistical analysis
Statistical analysis was carried out using SPSS software (version 21, SPSS Inc., Chicago, IL, USA). The data are presented as the mean ± S.E.M. Statistical analysis was performed with Student's t-test or oneway ANOVA, followed by Dunnett's test. All statistical tests were two-sided, and differences were considered to be significant at P≤0.05.

Expression of CENPO in NSCLC tissues and cell-lines
The expression of CENPO in NSCLC tissues was rstly evaluated by UALCAN database. The transcriptional level of CENPO was upregulated in both LUAD ( Figure 1A, P<0.001) and LUSC ( Figure 1E, P<0.001) when compared with normal lung tissues.High CENPO expression was associated with lymphnode metastasis and advanced stage in both LUAD ( Figure 1B and IC) and LUSC ( Figure 1F and  1G).Using the database of KM plotter, we found patient with higher CENPO was associated with worse overall survival (OS) in LUAD(HR=1.67, 95% CI=1.32-2.12, P=1.4×10-5, Figure 1D). However, no signi cant association was found in LUSC (P=0.35, Figure 1H).
The expression of 75 NSCLC and its corresponding adjacent normal lung tissues was examined by immunohistochemistry (IHC). The CENPO staining intensities in NSCLC was stronger than adjacent normal lung tissues ( Figure 1I and 1J). According to the IRS=2, we divided the expression of CENPOas high expression (n=36) and low expression group (n=39). We found high CENPO was more frequently in patients with advanced stage (P<0.001, Table 1). However, no signi cant association was found between CENPO expression and OS in our clinical species ( Figure 1K).Furthermore, the expression level of CENPO in NSCLC cell-lines was signi cant higher than normal lung epithelial cell-line EBC-1(P<0.05, Figure 1L).

Establishment of CENPO knockdown cells
Three shRNAs were used to establishment of CENPO knockdown cells. The knockdown e ciency was con rmed by qRT-PCR (Additional le 1: Figure S1A, B). At last,shRNA3 was chosen because the knockdowne cacy was the highest among three shRNAs in NCI-H1299. After 72h, the transinfection e cacy of shRNA1 was observed under uorescence microscope. As showed in Figure 2, it was demonstrates a >80% e ciency of infection and the normal cell condition (Figure 2A and 2D). Both qPCR and western blot suggested that transinfected with shCENPO signi cantly reduced CENPO mRNA and protein expression in both A549 ( Figure 2B and 2C) and NCI-H1299 cell-lines ( Figure 2E and 2F).

Reduced CENPO expression signi cantly inhibit cell proliferation, migration and invasive ability
Celigo cell counting assay was performed to evaluate the effect of CENPO on cell proliferation. We found that cells in shCENPO groups exhibit slower proliferation rate when compared with shCtrl groups in both A549 ( Figure 3A) and NCI-H1299 ( Figure 3B). Inwound healing assays, the observed time point of A549 was 24h and 72h ( Figure 3C), but for NCI-H1299, it was 8h and 36h ( Figure 3D). This was because the growth rate of two cell-lines was different. We found reduced CENPO dramatically suppressed the migration abilities of both cell-lines ( Figure 3C and 3D). The invasion capability of A549 and NCI-H1299 transfected with shCENPO was evaluated by matrigel invasion chamber assay. As expected, reduced shCENPO groups notably repress the invasion ability when compared with shCtrl groups in both A549 ( Figure 3E) and NCI-H1299 cell-lines ( Figure 3F).

Reduced CENPO expression induced cell apoptosisand result in cell cycle arrest
In ow cytometry assay, Annexin-FITC positive cells in shCENPO groups were signi cantly increased when compared with shCtrl groups in both A549 ( Figure 4A) and NCI-H1299 ( Figure 4B). Cell cycle analysis showed reduced CENPO expression increased S and G2 cell populationin A549. This increase in Sand G2-M phase cell populationwasaccompanied with a concomitant decrease of cell number in G0/G1 phase in A549 ( Figure 4C). In NCI-H1299, only increased G2 cell population accompanied withthe decrease of G1 cell populationwas observed ( Figure 4D).

Potential mechanism of CENPO in NCI-H1299
Firstly, Human Apoptosis Antibody Array analysis was applied to investigate the role of CENPO on apoptosis signaling pathway. We found reduced CENPO expression in NCI-H1299 result in the upregulation of BIM Caspase3 IGFBP-6 p21 and the downregulation of IGF-II, Survivin, XIAP.( Figure 5A and 5B).Western blot was also performed to investigate the effect of CENPO on PI3K/Akt/mTOR and cellcycle related protein mTOR, P-mTOR, CDK1, CDK6 and PIK3CA. In western blot analysis, we found H-1299 transinfected shCENPO reduced the expression of P-mTOR, CDK1, CDK6 and PIK3CA, no signi cant change was found in mTOR( Figure 5C).

Reduced CENPO inhibits NSCLC growth in vivo
To determine the effect of CENPO on tumor growth in vivo, male BALB/c nude mice were inoculated with NCI-H1299 transfected with shCENPO and shCtrl. Bioluminescent image of tumors suggested that the uprescence intensity in mice with shCENPO was lower than shCtrl groups ( Figure 6A and 6D). The tumor growth curves for mice was showed in Figure 6C, we found the tumor growth of shCENPO groups was slower than shCtrl groups. After the mice were sacri ced and tumors were Removed, we found the tumor size of shCENPO groups was smaller than shCtrl goup, what's more, no tumor were found in ve mice in shCENPO group ( Figure 6B), this was also agreed with the result of bioluminescent imaging. The tumor weight of shCENPO group was lighter than shCtrl group ( Figure 6F). IHC staining of Ki67 was applied and con rmed that the expression of Ki-67 was suppressed in shCENPO group ( Figure 6F). Taken together, Reduced CENPO inhibited tumorigenicity of NSCLC H1299 cell in nude mouse xenograft model

Discussion
The present study demonstrated that CENPO was up-regulated in NSCLC and was associated with unfavorable clinicopathologic feature such as TNM stage. Further in vitro and in vivo study demonstrated that reduced CENPO could attenuate the proliferation, migration and invasion ability of NSCLC cell-lines. Potential mechanism of CENPO in the progression of NSCLC was also explored; we found reduced CENPO attenuatemalignant phenotypevia promoted apoptosis signaling pathway, and affectedPI3K/Akt/mTOR pathway and cell-cycle related proteins.
Firstly, the expression level of CENPO in NSCLC and its relationship between clinicopathologic feature was investigated by TCGA database and our clinicalspecies from mRNA level and protein level. Both data suggested CENPO was frequently up-regulated in NSCLC when compared with normal lung tissues. Also, High CENPO expression was associated with malignant features such as TNM stage.However, the signi cant association between CENPO expression and OS was only found in LUAD from KM plotter. Further large sample size was necessary to verify the prognostic role of CENPO in NSCLC.
In vitro and in vivo study was further conducted to explore the role of CENPO in the progression of NSCLC. We found reduced CENPO inhibit the proliferation, migration and invasion ability of A549 and NCI-H1299. Reduced CENPO also induced apoptosis and result in cell cycle arrest. We found A549 and NCI-H1299 showed different biological behavior. NCI-H1299 growth more quickly than A549, as a result, in wound healing and assays, the observed time point of A549 was 24h and 72h, but for NCI-H1299, it was 8h and 36h. In cell cycle analysis, cell cycle arrest was happened in S and G2-M phase in A549, but in HCI-H1299, it was happened only in G2-M phase.
Further mechanism of CENPO in NSCLC was explored. Resisting Cell Death is a hallmark for cancer.Insu cient apoptosis was often observed in cancers (13).We investigated the effect of CENPO on apoptosis signaling pathway in NSCLC via Human Apoptosis Antibody Array. We foundapoptosis signaling pathway related proteinBIM, Caspase3, IGFBP-6 and p21 was upregulated and the IGF-II, Survivin and XIAP was downregulated in NCI-H1299 transinfected with shCENPO. BIM was a proapoptotic BH3-only protein, and caspase-3 was apoptotic protein. (14) Up-regulation of BIM and caspase-3induce apoptosis. Insulin-like growth factor (IGF)signalingalso involved in apoptosis. (15)IGFBP-6 increased apoptosis was related to nuclear localization and interaction withtranscription factors.(16)P21 exhibits anti-apoptotic activity owing to the inhibition of proteins involved in apoptosis.
(17)Survivin, a member of the inhibitor of apoptosis protein(IAP) family that inhibits caspases and blocks cell death.(18)XIAP, a member of the IAP, has been identi ed as a potent caspase inhibitor. (19)Downregulated Survivin and XIAP promote apoptosis. As a result, reduced CENPO induced apoptosis via upregulated apoptosis promoting proteins and downregulated apoptosis inhibitor.
In additional, several PI3K/Akt/mTOR pathway and cell-cycle related protein (mTOR, P-mTOR, CDK1, CDK6 and PIK3CA) were detected by western blot. P-mTOR, CDK1, CDK6 and PIK3CA was downregulated when reduced CENPO expression. PI3K/Akt/mTOR pathway is known to be frequently activated in several types of cancer and it plays a critical role in a number of key cancerous behaviors. Both PIK3CA, mTOR and p-mTOR belonged to this pathway. The PIK3CA gene encodes p110a, which is the catalytic subunit109 of PI3K, thus active PI3K-Akt signaling pathway. mTOR is a down-stream effector of PI3K-Akt signaling pathway. mTOR consists of two independent functional complexes, mTORC1 and mTORC2, which can be phosphorylated to p-mTOR regulating several cellular functions, including proliferation, differentiation, tumorigenesis, angiogenesis, autophagy, and apoptosis by activating ribosome biogenesis and protein synthesis. (20,21) In our study, we found reduced CENPO expression would downregulated the expression of PIK3CA and p-mTOR, suggesting CENPO may active PI3K/AKT/mTOR signaling pathway to promote cancer progression by avoiding apoptosis and other cellarbehaviors.
Cyclin-dependent kinases (CDKs) are critical regulatory enzymes that drive all cell cycle transitions. CDK1 emerged as a key determinant of mitotic progression. In metazoans, much of the control over cell cycle entry is elicited at the level of CDK4 and CDK6, which are responsive to numerous growth regulatory signals. In the early phase of G1, the D-cyclins interact with CDK4/6 to form complexes to phosphorylate pRB, which is a tumor repressor required for transition from the G1 to S phase. CENPO belones to Centrome protein, the de ciency of these proteins may affect the function of Centrome thus affected the cell cycle. (22) As previous studies reported, other Centrome protein such as CENPN knockdown (shCENP-N) cells showed depressed cellular proliferation by cell-cycle arrest at the G1 phase with up-regulation of p21Cip1 and p27Kip1 and down-regulation of cyclin D1, CDK2, and CDK4. (23) The mechanism of CENPO in cancer was insu cient, our research rstly demonstrated that CENPO may affect apoptosis related protein, cell cycle related proteins and PI3K/AKT/mTOR signaling thus regulated many malignant behaviors such as proliferation, apoptosis, cellcycle thus promoting the progression of NSCLC. However, whether these relationship had connected relationship or cross taking. As we know, PI3K/AKT/mTOR signaling pathway may regulate apoptosis, cell cycle,is CENPO activated PI3K/AKT/mTOR signaling pathway then reduced apoptosis, cellcycle, or CENPO could affected them respectively, or through other pathways. And their directly biding site and detailed regulation, these problem should be investigated in the further.

Conclusion
In conclusion, our study suggested CENPO gene overexpression in NSCLC. Reduced CENPOexpression substantially decreased the proliferation, migration and invasion ability, and promoting apoptosis and induces cell cycle arrest of NSCLC cell-lines. Preliminarymechanismresearchsuggested reduced CENPO could active apoptosis pathway, suppressing PI3K/AKT/mTOR pathway and down-regulation cell-cycle relatedproteins.Therefore, CENPO may be considered as an oncogene and it may be a new therapeutic target for the treatment of NSCLC. However, further studies,especially validation theprognostic role in larger clinical samples and explore the mechanism of CENPO were needed.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article.

Competing interests
The authors have no con icts of interest to declare.  (The rst layer asterisk which is right above the error bar representing comparison to normal group, and the above layers asterisk which were above asecondary line represent the comparison between corresponding groups that were covered by the line). I. Representative images with adjacent normal lung tissues and cancer tissues of IHC staining were shown. JThe IHC score for CENPO staining was higher in NSCLC and normal lung tissues.K. Kaplan-Meier plotter was used to analyze the protein expression data from IHC.K. Relative CENPO mRNA expressions in normal human bronchialepithelial cells (EBC-1) and NSCLC cell lines (A549, H1299 and SPC-A-1).*P < 0.05, **P < 0.01, and ***P < 0.001.