KPNA2 promotes renal cell carcinoma proliferation and metastasis via NPM

Abstract Karyopherin α2 (KPNA2), involved in nucleocytoplasmic transport, has been reported to be up‐regulated in tumorigenesis. However, comprehensive studies of KPNA2 functions in renal cell carcinoma (RCC) are still lacking. In this study, we aim to investigate the roles of KPNA2 in kidney tumour development. Our results showed that down‐regulation of KPNA2 inhibited the proliferation and invasion of kidney tumour cell cells in vitro, while the cell cycle arrest and cellular apoptosis were induced once KPNA2 was silenced. Repression of KPNA2 was proved to be efficient to repress tumorigenesis and development of kidney tumour in in nude mice. Furthermore, one related participator, NPM, was identified based on Co‐IP/MS and bioinformatics analyses. The up‐regulation of NPM attenuates the efficiency of knockdown KPNA2. These results indicated that KPNA2 may regulate NPM to play a crucial role for kidney tumour development.

nucleocytoplasmic transport 16,17 . KPNA2 is essential for proper nuclear localization and multiple functions of NBS1, and the interaction between KPNA2 and NBS1 is crucial for the DNA repair role process 18,19 . The abnormal expression of KPNA2 exhibited a considerably changed activity in DSB repair, which is proved critical for the DNA damage-activated cell cycle checkpoint. Apart from this, KPNA2 has also been identified to mediate nuclear transport of tumour suppressors 17,20 . Recent studies have demonstrated the association between the up-regulation of KPNA2 and various types of malignancies [21][22][23][24] . These results suggested the effect exerted by KPNA2 in tumour formation and progression is through cell differentiation, proliferation and apoptosis. Although accumulating evidence confirmed the participating of KPNA2 in the progression of tumour, little is known about the oncogene effects of KPNA2 in ccRCC, as well as its potential effective function for suppression of ccRCC development.
As a well-known nucleolar phosphoprotein, NPM1 has been found to be strongly correlated with cell proliferation and cancer pathogenesis. This gene has oncogenic and tumour-suppressing functions through the frequent overexpression or genetic modification [25][26][27][28] . The mutation of NPM1 has been proved to be related to leukaemia and lymphoma 29,30 . NPM1 is overexpressed in a variety of human solid tumours including tumours of colon 31 , ovary 32 and prostate and other tumours 33,34 . The specific contribution of overexpressed NPM1 to cancer development is not fully understood, but it may arise from multiple factors. Inhibition of NPM1 effectively suppressed the viability of tumour cell lines [35][36][37] , making it as one of the hopeful targets for cancer treatment 28 .
In this study, we found the expression of KPNA2 was signifi- Laboratories, Logan, UT, USA) supplemented with 10% foetal bovine serum (FBS, Gibco), 100 µg/ml streptomycin and 100 µg/ml penicillin. Caki-1 cells were cultured in RPMI 1640 medium with 10% (v/v) foetal calf serum (FCS) and 100 U/ml penicillin. All cells were incubated in an incubator at 37℃ with a humidified atmosphere of 5% CO 2 . When the cells in the culture flask reached 80% confluence, the culture solution was discarded and washed once, and 2.5g/L trypsin was added at a dilution of 1: 5.
KPNA2 siRNAs were inserted into the GV610 lentivirus core vector containing an enhanced green fluorescent protein (Genechem, Shanghai, China; Figure S1) using the Fermentas T4 DNA Ligase (Fermentas, Burlington, Canada) according to the manufacturer's instructions. The positive clones were sequenced using the identified primers, and clones that are consistent with the target sequence were selected. Lentiviral vector without siRNA insertion was used as a negative control.

| Cell transfection
To investigate the role of KPNA2 in human renal clear cell carcinoma cell, 786-O and ACHN were infected with lentiviruses expressing either shCtrl or shKPNA2. In brief, cells were seeded into 6-, 12-or 24-well plates, and when the cells grew up to about 30%, the lentiviral vector was added to the cells. The cultures were abandoned, and normal culture medium was added 12 hours later. When the positive rate of GFP was up to 70%-80% and the cell density was about 80%, the cells were collected and used in the following experiments. The percentage of cells that was GFP-positive cells represented the lentiviral infection efficiency, which was examined using a fluorescence microscope.
The full-length coding sequence of the human KPNA2 was cloned into the lentiviral vector GV610 lentivirus core vector with Cterminal 3 × FLAG tag (Jikai, Shanghai, China). The KPNA2-encoding virus was then transfected into ACHN cell line. The full-length coding sequence of the human NPM1 was cloned into the lentiviral vector GV610 lentivirus core vector containing EGFP. Cells were cultured in 96-well plates. And when the cells grow up to about 30%, lentiviral vectors with target sequence were added to the cells. The cultures were abandoned, and normal culture medium was added 12 hours later. The expression of green fluorescent protein (GFP) was observed under fluorescence microscope. When the positive rate of GFP was up to 70%-80% and the cell density was about 80%, the cells were collected and used in the following experiments. The primers used were as follows: KPNA2, 5'-TGTGGTAGATGGAGGTGC-3' (forward) and reverse, 5'-GAGCCAACAGTGGGTCA' (reverse). The experiments were repeated three times.

| Western blotting for protein expression analysis
Total cellular proteins were harvested and separated by SDS-PAGE gels and then transferred onto PVDF membranes (Millipore).
Equal protein loading was verified by the Bradford assay. The primary antibodies used in Western blot were the same as those in 'Immunocytochemistry'. Membranes were blocked for 30 min at room temperature with 5% non-fat dry milk in TBS containing 0.05% Tween-20 (TBST), probed with the primary antibodies at a 1:1000 dilution for 1 h at room temperature and washed four times with TBST. The secondary anti-mouse/rabbit IgG-HRP (DAKO) was added for 1 h at room temperature. After extensive rinsing, immunoreactive protein bands were visualized with the ECL detection system (Santa Cruz, USA) and subsequently exposed to film. Densitometric assay was performed using Quantity one software (Bio-Rad, USA).
The specific primary antibodies were purchased from the following

| Cell growth assays
Cell growth was examined using Celigo (Nexcelom, Lawrence, MA) with EGFP. Kidney tumour cells in each experimental group in logarithmic growth phase were made into cell suspensions using trypsin.
Cells were then seeded in 96-well plates at a density of 1 × 10 3 or 2 × 10 3 cells/well. Each group had three plates, and the culture system was 100 μl per plate. The cells were cultured at 37℃. From the second day after plating, the plates were scanned once a day by Celigo (Nexcelom, Lawrence, MA) for 5 days.

| MTT assay
Cell growth status was determined using a routine MTT assay.
Briefly, kidney tumour cells infected with lentiviruses either shCtrl or shKPNA2 were made into cell suspensions. The cells in the logarithmic phase were then seeded in 96-well plates at a density of 2 × 10 3 cells/well. The cell growth data were determined each day for 5 continuous days. 20 μL of 5 mg/mL MTT was added to the wells 4 hours before removing the media at each day. And 100 μL dimethyl sulphoxide (DMSO) was then added to dissolve the resultant MTT formazan. The optical density value was measured at 490/570 nm using a microplate spectrophotometer (Tecan Infinite, Hombrechtikon, Switzerland). The experiments were repeated three times, and the data were statistically analysed and the cell growth curves were drawn.

| Cell cycle analysis
After infected with target virus, cell cycle distribution was investigated to evaluate the effect of KPNA2 on kidney tumour cell.
Briefly, the cells were made into cell suspensions using trypsin and collected in a 5-mL centrifuge tube. The cells were centrifuged at 1300 rpm for 5 minutes, and then, the supernatant was discarded.

| Cell apoptosis
The apoptosis of transfected cells was examined by the annexin V-APC assay followed by flow cytometry. Briefly, the cells infected by lentiviruses were cultured for 3 days. Then, the cells were collected and washed with PBS. The cells were suspended using 200 μL 1 × binding buffer at a final density of 1 × 10 6 /ml. Then, 10 μL annexin V-APC (eBioscience, Waltham, MA) was added to 100-μL cell suspensions in a dark place at room temperature for 10 to 15 minutes.
Then, the percentage of the apoptosis of the cells was examined by flow cytometer (Millipore, Germany).

| Wound healing assay
Kidney tumour cells in in the logarithmic phase were seeded on a collagen type 1-coated, 6-well plate and grown overnight to 80% confluence. After overnight serum starvation, a wound was created using a P200 pipette tip. The culture wound was photographed at time 0, 4 and 8 h later, and the rate of closure was assessed.

| Transwell assay
Migration was performed using the transwell filter chambers
Cells were then incubated with 1% normal goat serum (Solarbio) for 1 h and incubated with primary antibodies overnight at 4℃. Then, cells were washed and incubated with an Alexa Fluor® 594 secondary antibody for 1 h at room temperature. Finally, nuclear staining was performed with DAPI (Beyotime) for 10 min. Representative images were acquired using a confocal microscope (Olympus, Japan). Survival differences among groups were estimated using the R package 'survival'. The Kaplan-Meier curve was constructed, and a logrank test was used to determine differences among overall survival and progression-free survival according to the KPNA2 mRNA levels in this study.

| KPNA2 is up-regulated in ccRCC
To investigate the role of KPNA2 related to the development of

| Knockdown of KPNA2 inhibits kidney tumour cell growth
The  Figure 2E and 2F). The results showed that 786-O cells and ACHN cells were arrested at different stages.

| Knockdown of KPNA2 promotes kidney tumour cell apoptosis
Subsequently, we investigated whether the apoptotic cell death could be affected by the KPNA2. Then, flow cytometry was performed. As shown in Figure 3A and 3B A significant increase in apoptosis phenomenon was observed in shKPNA2 cells. Both in 786-O cells and in ACHN cells, apoptosis was found in less than 5% cells infected with lentiviruses without shKPNA2, while an apoptosis rate over 10% was observed in KPNA2 knockdown cells. The results indicated that knockdown of KPNA2 promotes kidney tumour cell apoptosis.
F I G U R E 1 KPNA2 is correlated with the development of ccRCC. A, Gene expression analysis showed that KPNA2 was significantly up-regulated in clinical stage III/IV patients compared with stage I/II patients. B, Kaplan-Meier analysis showed that the expression of KPNA2 was significantly associated with poor overall survival rate in ccRCC patients (P-value <0.05, log-rank test). C, The high expression of KPNA2 showed a poor progression-free survival (PFS) rate in ccRCC patients (P-value <0.05, log-rank test)

| Knockdown of KPNA2 inhibits kidney tumour cell migration and invasion
Recent studies also indicated that high expression of KPNA2 was correlated with the development and progression of several tumours 21,39 . We further assessed the effect of KPNA2 on cell migration and invasion. When KPNA2 was down-regulated, the migration of cells increased slightly compared with the control cells in the following 8 hours ( Figure S3 and S4). However, the invasion ability was significantly reduced when KPNA2 was down-regulated (p<0.05, Figure 4). These in vitro data demonstrated that KPNA2 has an impact on metastasis of kidney cancer cells.

| Repression of KPNA2 reduced the development of kidney cancer in vivo
To confirm the oncogenic activity of KPNA2 in vivo, we established a BALB/c nude mouse model using ACHN cells with shKPNA2 or shCtrl. After 4 weeks, ACHN cells with shKPNA2 or shCtrl were microinjected into right armpit of female nude mice and the tumour volume was measured every 7 days until day 49. The luciferase was employed to measure the tumour area, as well as tumour metastasis and severity ( Figure 5A and 5B). Bioluminescence imaging in living mice clearly showed that EGFP signals were significantly higher in the NC group than those in the shKPNA2 group. Statistical analysis demonstrated that the tumour volume and weight of mice treated with the shKPNA2 were significantly decreased relative to those of NC (p<0.05, Figure 5C-E). Mice in the shKPNA2 group exhibited an obviously repressed development of metastasis than those in the NC group. These data proved that knockdown of KPNA2 is sufficient to repress the development of tumour.

| NPM1 was one of the targets of KPNA2
To screen out the interacting proteins of KPNA2, we conducted co-  (Table S1; TableS2). 209 proteins were identified with relatively high expression level in KPNA2 overexpression cells (Table S3). Western blotting analysis was performed to evaluate the expression level of four candidate KPNA2 interaction proteins from Table S3 ( Figure 6C).
Co-IP results showed a specific interaction between KPNA2 and NPM1( Figure 6C). IF staining further validated a considerable degree of colocalization for KPNA2 and NPM1 in ACHN cells. The expression of two proteins was colocalized in the nucleus ( Figure 6D).
Then, Western blotting of NPM1 with KPNA2 knockdown in ACHN cells was performed. We found that NPM1 expression was consistent with KPNA2 ( Figure 6E-F). Also, c-Myc was down-regulated after KPNA2 knockdown ( Figure 6E). A previous study indicates that KPNA2 can transport c-Myc into the nucleus, which could trigger tumorigenicity of cells 39 , and c-Myc can directly bind to the NPM1 promoter to induce its expression 40 . Further, the expression of KPNA2 is positively correlated with NPM1 in kidney cancer ( Figure 6G). Also, we found that correlation between Myc and KPNA2, Myc and NPM1 was significantly positive in KIRC ( Figure 6H-I). Furthermore, by using the String database(https://www.strin g-db.org/), we found that KPNA2, NPM1 and Myc interact closely with each other ( Figure 6J). So, KPNA2 may regulate NPM1 via c-Myc to promote the proliferation of kidney cancer cells.

| KPNA2 promotes kidney cancer cell proliferation and migration by targeting NPM1
To assess whether NPM1 contributes to the proliferation and migration progression of kidney cancer through the regulation of KPNA2, we transfected KPNA2-targeting siRNA into ACHN cells with forced expression of NPM1. As shown in Figure 7, overexpression of NPM1 attenuated the repression of silencing KPNA2 on cell proliferation.
The proliferation rate significantly reduced after silencing KPNA2, while increasing the expression of NPM1 led to the similar proliferation rate compared with control cells, which was also indicted by MTT assay results. Furthermore, overexpression of NPM1 also promoted the migration and invasion of cancer cells even though KPNA2 was repressed. These data suggested that NPM1 could rescue phenotypic deletion caused by KPNA2 knockdown in ACHN cells.

| DISCUSS ION
As one of the major roles involved in nucleocytoplasmic transport process, KPNA2 has been proved to be related to proliferation, Cell cycle was abrupted after the repression of KPNA2 with siRNA, which might eventually be helpful to inhibit tumour cell growth in ccRCC. Tumour invasion is also one of the key steps leading to metastasis and poor prognosis 47 , which could arise from several consecutive mutation events and epigenetic elements. We discovered that invasion and migration in both cancer cells were inhibited after silencing KPNA2.
Based on the Co-IP/MS analysis, we identified a key protein, NPM1, that could interact with KPNA2. Even though down-regulation of KPNA2 could significantly repress cancer cell proliferation and further invasion ability, we found that overexpression of NPM1 could promote cell proliferation, as well as the migration ability and invasion of cancer cells in some degree. These results firstly proved the cooperation of KPNA2 and NPM1 in tumorigenesis.
As a well-known nucleolar phosphoprotein, NPM1 has been found to be strongly correlated with cell proliferation and cancer pathogenesis. This gene has oncogenic and tumour-suppressing functions through the frequent overexpression or genetic modification [25][26][27] . Mutation of NPM1 had been proved to be related to leukaemia and lymphoma 29,30 . Furthermore, NPM1 is overexpressed in many types of major human solid tumours including tumours of colon 31 , ovary 32 and prostate and other tumours 33,34 . Moreover, the oncogenic c-Myc can directly bind to the NPM1 promoter to induce its expression in leukaemia 40 . Previous study also proved that overexpression of KPNA2 can transport c-Myc into the nucleus, which could trigger tumorigenicity of cells 39 . So, we hypothesize that the overexpression of KPNA2 in ccRCC triggers the up-regulation of c-Myc, which further induced the expression of NPM1, and eventually contributes to tumorigenesis. In addition, the detailed mechanism of KPNA2 and NPM1 as an oncogenic factor in cancer cells should be evaluated in future studies.
In this study, we evaluated the expression level of KPNA2 in ccRCC patients from different stages. KPNA2 is overexpressed in the late stage of ccRCC, and its up-regulation is associated with poor prognosis. Significantly higher expression of KPNA2 was found in the late stage, while higher KPNA2 is significantly associated with poor survival rate. Next, in in vitro and vivo analysis, our results indicated that KPNA2 may regulate NPM1 to promote the progression of kidney cancer.

ACK N OWLED G EM ENTS
This work was supported by grants from the Natural Science

Foundation of Fujian Province (2019J01153) and Startup Fund for
Scientific Research, Fujian Medical University (2019QH1053).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.