RPS15A promotes the proliferation and migration of nasopharyngeal carcinoma


 Background: Nasopharyngeal carcinoma (NPC) is an epithelial squamous cell carcinoma arising from the nasopharyngeal mucosal lining. Ribosomal protein S15a (RPS15A) plays vital role in protein translation and was recently reported to be an oncogene in numerous tumor types. However, its biological role in NPC is remain largely unclear.Methods: In this study, we explored the expression of RPS15A in NPC tissues by immunofluorescence histochemical staining (IHC) staining of human tissue microarray. The C666-1 and CNE-2Z NPC cell lines with RPS15A depletion were used to investigate the effects of RPS15A on NPC cell proliferation, migration and apoptosis. In vivo tumor growth of NPC cells was observed by subcutaneous xenograft mice model. The potential mechanism was explored by Human Apoptosis Antibody Array analysis and WB experiments. Results: RPS15A was significantly up-regulated in NPC and RPS15A knockdown remarkably suppressed NPC cells proliferation, migration and induced cell apoptosis. Moreover, RPS15A silencing also impaired tumor growth of xenograft mice. Further Human Apoptosis Antibody Array analysis indicated that depletion of RPS15A could promote several apoptosis-related proteins expression, and results of WB experiments confirmed the inhibition of PI3K/AKT pathway. Conclusion: RPS15A knockdown suppressed proliferation, migration and increased apoptosis of NCP cells by inhibiting PI3K/AKT signaling pathway. RPS15A may serve as a promising therapeutic target for NPC patients.


Introduction
Nasopharyngeal carcinoma (NPC) is an epithelial squamous cell carcinoma arising from the nasopharyngeal mucosal lining [1]. Compared to other cancers, NPC is an uncommon tumor type, which is distinctly different from other epithelial head and neck tumors and has unique distribution that mainly focus on East and Southeast Asia [2,3]. Epidemiological reports suggested that incidence of NPC has declined gradually worldwide in the past decades [4,5]. It may owe to the signi cant advancements in imaging diagnosis technology, radiotherapy delivery, and systemic chemotherapy for NPC patients [6].
Moreover, developments of novel therapeutics, such as immune checkpoint therapies and targeted therapies have obviously improved the prognosis of NPC patients with recurrence or metastasis [7][8][9].
However, several questions about pathogenesis and clinical management on NPC still need to be addressed [2]. A better understanding of biological mechanisms underlying NPC pathogenesis may allow the identi cation of biomarkers related to NPC progression and management.
Ribosomes is one of organelles of eukaryotes, consisting of a small 40S subunit and a large 60S subunit, responsible for protein synthesis [10]. It's found that regulation of cell proliferation and differentiation by ribosome biogenesis and protein translation is abnormal in tumor cells [11]. The mutations of oncogenes and cancer suppressed genes often stimulate cell growth and proliferation by enhancing ribosome biogenesis in cancer cells, thereby driving cancer occurrence [12]. Ribosomal protein S15a (RPS15A) encodes a component of 40S ribosomes subunits, plays vital role in protein translation [13]. Recently, accumulating evidences showed the upregulation of RPS15A in a plethora of cancers, including hepatocellular carcinoma [14], pancreatic cancers [15], breast cancer [16], and colorectal cancer [17], etc. RPS15A knockdown inhibited proliferation of hepatic cancer cells [18], lung cancer cells [19] and osteosarcoma cells [20] in vitro, kidney cancer cells growth in vitro and in vivo [21], as well as inducing cell apoptosis in glioblastoma [22]. However, the biological functions of RPS15A in NPC is largely unknown.
In this study, we identi ed the up-regulation of RPS15A in NPC by bioinformatics analysis and IHC staining of human tissues. Moreover, we explored the effects of RPS15A on proliferation, migration and apoptosis of NPC cells by RPS15A stable knockdown cell models. Additionally, we preliminarily investigated the apoptosis-related mechanism of RPS15A-induced malignant phenotypes of NPC cells by Human Apoptosis Antibody Array.

Materials And Methods
Clinical tissue samples and immuno uorescence histochemical staining (IHC) staining Human tissue chip (Cat No. HNasN132Su01) containing 107 of NPC tissues and 30 of para-carcinoma normal tissues was purchased from Shanghai Outdo Biotech Company (Shanghai, China) and was used to detect the expression of RPS15A. The detailed pathological characters and the informed consents from patients were collected. This study was approved by the Ethics Committee of Kunming Medical University. For the IHC staining of RPS15A, brie y, all the xed tissues were rstly dewaxed followed by rehydrating. Then the tissues were incubated with primary antibody anti-RPS15A (1:100, Lot No.PA5-51314, Invitrogen) at 4°C overnight following antigen repair and blocking. The secondary HRP-conjugated goat anti-rabbit IgG (1:400, Lot No. ab97080, Abcam) were then incubated with tissue slides at 37°C for 1 h. These slides were nally colored by diaminobenzene (DAB), counterstained by hematoxylin, dehydrated by alcohol, transparent with xylene, and sealed by neutral gum, respectively. The results of IHC staining were reviewed by two independent histopathologists and scored by percentages of positive staining cells (1, 0~24%; 2, 25~49%; 3, 50~74%; 4, 75~100%) and the staining intensity (0, no staining signals; 1, light yellow; 2, pale brown; 3, seal brown). The nal IHC scores were determined as product of percentages of positive staining cells and the staining intensity: 0 score (-), 1-4 scores (+), 5-8 scores (++), 9-12 scores (+++) [23].
Transfection e cacy was determined by observing the GFP expression.
Cell proliferation assay Cell viability was assessed by MTT assays. In brief, the NPC cells were transfected with shCtrl or shRPS15A lentivirus until grew at 85% con uence. Cell were then harvested and plated into 96-well plates at a density of 2 × 10 3 cells per well. 24 h after inoculation, 20 μL of MTT solution (5 mg/mL) was added into each well 4 h before termination of culture and incubated for 4 h. After that, 100 μL of DMSO was added to stop reaction. Finally, the optical density (OD) value at 490 nm was determined by microplate reader (Thermo, USA) and the growth curve was plotted based on the OD value.

Colony formation assay
The NPC cell lines transfected with indicated lentivirus for 5 days were harvested for colony formation assays. The transfected cells were counted and seeded into 6-well plates at a density of 900 cells per well. After 8 days of culture, the cell colonies were xed by 4% paraformaldehyde for 30 min and stained by GIEMSA for 5 min. Then the colony numbers were determined under microscope.

Cell apoptosis analysis
Cell apoptosis was examined by ow cytometry. Firstly, the NPC cell lines were transfected with indicated lentivirus and cultured until the cell fusion reached 85%. Cells were then collected and re-suspended for Annexin V-APC (eBioscience, USA) staining. 10 min of incubation away from light, cells were washed three times with PBS followed by quanti cation of positive staining cells by ow cytometer (Millipore, USA).

Cell migration analysis
Cell migration was evaluated by Wound-healing assays and Transwell assays. In brief, the NPC cell lines transfected with indicated lentivirus were harvested when grown at 80% con uence. For wound-healing assays, cells were then re-suspended, counted and plated into 96-well plates at a density of 5 × 10 4 cells per well. The next day, scratches were formed using a scratch tester to push from bottom center of the 96well plate to top. After removing of exfoliated cells by serum-free medium washing, cells were then cultured with 0.5% FBS-containing medium until indicated time points. Cellomics (Thermo, USA) was used to determine migratory distance of cells in each group, and the migration rate was calculated based on the migratory distance.
For transwell assays, the required numbers of chambers were placed into an empty 24-well plate followed by addition of 100 μL cell suspensions. Besides, 600 μL of 30% FBS-containing medium was added into the lower chambers. After incubation for 24 h, the non-migrating cells were removed with cotton tip and then the cells were xed with 4% paraformaldehyde for 30 min and stained with Giemsa for 5 min. The migratory cells were determined under uorescence microscope (Olympus, Japan).

Mice xenograft Model
10 of BALB/c nude mice (female, four-weeks-old) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd (Beijing, China). Animal procedures were approved by Animal Care Committee of Kunming Medical University (No. kmmu 20211172). The xenograft models were established by subcutaneous injection of C666-1 cells (4×10 6 per mice) transfected with indicated lentivirus. One week after injection, the tumor sizes were monitored twice weekly using a Vernier caliper, and the tumor volumes were calculated as follows: V=π/6×L×W 2 (L, represents the longest diameter; W, represents the shorter diameter). At the end of this experiments, all mice were sacri ced by injection of pentobarbital sodium as described previously [24] and the tumor were removed for weighting, photographs and Ki-67 staining.

Human Apoptosis Antibody Array
The Human Apoptosis Antibody Array (Cat No. ab134001, Abcam) assay was performed to reveal the expression alterations of apoptotic proteins. Brie y, the C666-1 cell transfected with shCtrl or shRPS15A lentivirus were prepared and lysed by lysis buffer. Cell lysates were then blocked using blocking buffer followed by incubation with the biotin-conjugated anti-cytolines at 4°C overnight, and streptavidin-HRP at room temperature for 2 h, respectively. Finally, the pixel density on the membrane was evaluated by enhanced ECL and quanti ed by ImageJ software.

Bioinformatics analysis
The independent dataset GSE12452 containing 10 of normal nasopharyngeal tissue and 31 of NPC tissues were available on GEO database (https://www.ncbi.nlm.nih.gov). RNA-Seq data were expressed as Fragments Per Kilobase Million (FPKM). The differentially expressed genes (DEGs) between normal tissues and tumor tissues were determined by following criteria: |Fold Change| ≥ 1.5 and p value < 0.05. RPS15A was identi ed as a signi cant upregulated gene in NPC tissues compared to that in normal tissues.

Statistical analysis
Each experiment in this study was performed in triplicate. Graphpad Prism 8.04 software was used to perform statistical analyses and graph. Data were shown as mean ± SD. The two-tailed Student's t-test was used to determine p values between two groups. For multiple groups, the p values were determined by one-way analysis of variance (ANOVA). A value of p<0.05 was considered statistically signi cant.

Results
Identi cation of RPS15A as a critical gene in NPC progression Based on the gene microarray data of 10 normal nasopharyngeal samples and 31 NPC samples from the GEO database GSE12452 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE12452), a total of 3334 DEGs were identi ed, with 1755 up-regulated mRNAs and 1579 down-regulated mRNAs (|Fold Change| |≥1.5, FDR <0.05) (Fig. 1A). Among them, RPS15A were signi cantly up-regulated NPC tissues compared to normal nasopharyngeal tissues (P=0.0014 and log |FC| =0.68) (Fig. 1B). Moreover, IHC staining of human tissue chips showed that RPS15A was mainly presented in the NPC tissues, especially expressed strongly in metastatic NPC tissues, while weak staining for RPS15A was observed in normal nasopharyngeal tissues (Fig. 1C, Table 1). We further detected the correlation of RPS15A expression with clinicopathological characteristics of NPC patients. Intriguingly, both Mann-Whitney analysis and Spearman correlation analysis con rmed that RPS15A levels were positively associated with cervical lymph node metastasis ( Table 2 and Table 3). Collectively, these results suggested that RPS15A played a vital role in the progression of NPC.

Construction of RPS15A knockdown stable cell lines in NPC cells
In order to explore the speci c role of RPS15A in NPC, C666-1 and CNE-2Z cell lines were selected to construct RPS15A knockdown cell model. Brie y, three shRNA sequences targeting RPS15A gene were designed and applied to interfere NPC cells. Among three shRNAs, shRPS15A-1 and shRPS15A-2 demonstrated high e ciency of EIF3B knockdown, especially shRPS15A-1 ( Fig. 2A). Moreover, Fig.  2B revealed that RPS15A was smoothly knocked down using lentivirus mediated RPS15A shRNA (shRPS15A-1) in C666-1 and CNE-2Z cells. The mRNA and protein levels of RPS15A in C666-1 and CNE-2Z cells were visibly decreased after infection with shRPS15A (Fig. 2C-D). Collectively, these results indicated that RPS15A knockdown was constructed successfully in C666-1 and CNE-2Z cell lines and shRPS15A-1 could be used to perform further cell function tests.

RPS15A knockdown repressed NPC cell viability
To determine the effect of RPS15A knockdown on cell viability in vitro, cell proliferation was performed on C666-1 and CNE-2Z cells by using the MTT assay. Our results showed that RPS15A knockdown signi cantly suppressed cell proliferation of C666-1 and CNE-2Z cells (Fig. 3A). Moreover, Lv-shRPS15A infected NPC cells formed much fewer colonies compared with those obtained with Lv-shCtrl infected cells (Fig. 3B). Furthermore, ow cytometry with Annexin V-APC single-staining method showed that RPS15A knockdown boosted cell apoptosis in C666-1 and CNE-2Z cells (Fig. 3C). These results demonstrated that RPS15A knockdown suppressed cell viability both in C666-1 and CNE-2Z cells.

RPS15A knockdown repressed NPC cell migration
The role of RPS15A in cell migration of C666-1 and CNE-2Z cells was further investigated by wound healing assay and Transwell assays. In wound healing assays, the C666-1 and CNE-2Z cells transfected with shRPS15A lentivirus displayed impairment of migration ability at 48 h compared to cells transfected with shCtrl lentivirus (Fig. 4A). Moreover, Transwell assays also showed that RPS15A knockdown attenuated the migration ability of C666-1 and CNE-2Z cells (Fig. 4B). Collectively, these results suggested that RPS15A knockdown inhibited the migration capacities of NPC cells in vitro.

RPS15A knockdown inhibited tumor growth in nude mice
Given our in vitro results showed that RPS15A knockdown could suppress NPC tumor cell phenotypes, we further investigated the effects of down-regulated RPS15A on NPC tumor growth in vivo using xenograft mouse model. Brie y, RPS15A knockdown C666-1 cells were subcutaneously injected into nude mice (n=5), and down-regulated RPS15A signi cantly decreased tumor growth in vivo compared with shCtrl group (n=5) (Fig. 5A). 27 days after injection, mice were sacri ced, and tumors were separated to weigh. As shown in Fig. 5B, the size of tumor in shRPS15A group were signi cantly smaller than those in the shCtrl group. Moreover, RPS15A knockdown diminished tumor weight (Fig. 5C). Consistently, the expression of Ki67 protein was signi cantly inhibited by shRPS15A, suggesting that the proliferation ability of tumor cells was weakened (Fig. 5D). Taken together, these data indicated that RPS15A knockdown inhibited NPC tumor growth in vivo.

RPS15A knockdown regulated the expression levels of proteins related to biological function of NPC cells
To further unveil the mechanisms behind RPS15A regulating the malignant behaviors of NPC cells, we rst carried out a Human Apoptosis Antibody Array in C666-1 cells. As shown in Fig. 6A-B, RPS15A knockdown increased the expression of the pro-apoptotic elements CD40, CD40L, cytoC, DR6, FasL, HSP60, IGFBP-3, p21 and TNF-β. These data further imply that RPS15A knockdown contributed to apoptosis of NPC cells. Additionally, we also found that RPS15A knockdown signi cantly down-regulated the protein levels of P-AKT, CCND1, CDK6 and PIK3CA (Fig. 6C). Taken together, these results suggested that RPS15A depletion had an inhibitory effect on the progression of NPC.

Discussion
NPC is a high incidence of malignant squamous cell carcinoma in southern China, southeast Asia and north Africa, threatening the health of many people [25]. Despite great progress in radiotherapy and chemoradiotherapy for NPC, the e cacy of locally advanced NPC remains suboptimal [26]. At present, efforts have been made to identify biomarkers that can diagnose the risk of NPC patients or novel targeted therapy that can improve the prognosis of patients [25]. Therefore, it is necessary to explore the mechanics of NPC progression in depth. In recent years, the relationship between ribosomal protein family members and tumor progression has attracted much attention. There is considerable evidence indicated that RPS15A is abnormally high expressed in various types of human cancers, including liver cancer, colorectal cancer, lung adenocarcinoma, breast cancer, and osteosarcoma [16,18,20,27,28]. In this study, we highlighted that RPS15A is abundantly expressed in NPC and positively correlated with cervical lymph node metastasis. These results suggested that elevated expression of RPS15A predicts the deepening of tumor malignancy, which may be used as a novel diagnostic biomarker for NPC with important clinical signi cance.
Moreover, the functions of RPS15A in various biological processes have been clari ed, such as being responsible for the regulation of cell division and participating in tumor progression [17,29]. In particular, the role of RPS15A in tumor progression has been increasingly recognized [15,19,22,[30][31][32][33]. Therefore, RPS15A is currently considered as a promising target for cancer therapy in the design of potent anticancer drugs. Consistently, the present study demonstrates a key driving role of RPS15A in the proliferation and migration of NPC. Moreover, we further revealed that knockdown of RPS15A signi cantly inhibited the tumorigenic ability of NPC cells in mice. The above ndings strongly reveal that RPS15A plays an extremely important role in promoting NPC. Furthermore, our data suggested that knockdown of RPS15A can induce apoptosis in NPC cells. Meanwhile, knockdown of RPS15A resulted in abnormal expression of a series of apoptosis-related factors, such as up-regulation of CD40, CD40L, cytoC, DR6, FasL, HSP60, IGFBP-3, p21 and TNF-β. Therefore, we suspected that RPS15A regulates the progression of NPC by activating these apoptosis-related factors to initiate apoptotic signaling pathways.
As we all known, PI3K/AKT signaling pathway has been widely reported to be one of the most common and important pathways for cancers [34][35][36]. The pathway includes multiple core components and is controlled at multiple levels. In this study, we found that RPS15A knockdown could down-regulated the expression of p-AKT and PIK3CA. Additionally, knockdown of RPS15A signi cantly reduced the expression of CCND1/ /CDK6 at the protein level. CCND1 and CDK6 are cell cycle promoter that induces cell proliferation in tumors [37]. Taken together, these results suggested that RPS15A can induce NPC cell progression through the PI3K/AKT signaling pathway and CCND1/CDK6. Of course, the speci c molecular mechanism is worthy of our further study.

Conclusion
In this study, we revealed the expression level and role of RPS15A in NPC. Our data showed that RPS15A is signi cantly up-regulated in NPC. Moreover, knockdown of RPS15A inhibited the proliferation, migration and tumorigenic capacity of NPC cells. In summary, RPS15A may enhance NPC progression by PI3K/AKT signaling pathway, which may be a potential therapeutic target to prevent NPC progression.   The mRNA levels of RPS15A in C666-1 and CNE-2Z cells was detected by qPCR to assess the knockdown e ciency. (D) The protein levels of RPS15A in C666-1 and CNE-2Z cells was detected by western blot to assess the knockdown e ciency. Results were presented as mean ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001.  was detected by immunohistochemistry. Results were presented as mean ± SD. ** P < 0.01, *** P < 0.001.

Figure 6
The effects of RPS15A knockdown on the proteins related to biological function of NPC cells (A) The levels of apoptosis-related proteins in C666-1 cells infected with shCtrl and shRPS15A were measured using a human apoptosis antibody array. (B) Protein expression was visualized by R studio and presented in gray value. (C) Several cancer-related protein was detected by WB. GAPDH was used as the loading control. Results were presented as mean ± SD. * P < 0.05, ** P < 0.01.