The DEAD-box RNA helicase 51 controls non-small cell lung cancer proliferation by regulating cell cycle progression via multiple pathways

The genetic regulation of cell cycle progression and cell proliferation plays a role in the growth of non-small cell lung cancer (NSCLC), one of the most common causes of cancer-related mortality. Although DEAD-box RNA helicases are known to play a role in cancer development, including lung cancer, the potential involvement of the novel family member DDX51 has not yet been investigated. In the current study we assessed the role of DDX51 in NSCLC using a siRNA-based approach. DDX51 siRNA-expressing cells exhibited a slower cell proliferation rate and underwent arrest in S-phase of the cell cycle compared with control cells. Microarray analyses revealed that DDX51siRNA expression resulted in the dysregulation of a number of cell signalling pathways. Moreover, injection of DDX51 siRNA into an animal model resulted in the formation of smaller tumours compared with the control group. We also assessed the expression of DDX51 in patients with NSCLC, and the data revealed that the expression was correlated with patient age but no other risk factors. Overall, our data suggest for the first time that DDX51 aids cell cancer proliferation by regulating multiple signalling pathways, and that this protein might be a therapeutic target for NSCLC.

promoting cell proliferation in NSCLC, the roles of other members of the family are more elusive. For example, DDX51 is involved in regulating RNA metabolism, and in particular in the maturation of pre-RNAs 10 . However, the clinical importance of this protein in the context of NSCLC has not been assessed previously.
In the current study we used a siRNA silencing approach to investigate the role of DDX51 as a transcriptional regulator in NSCLC for the first time. DDX51siRNA H1299 cell cultures exhibited a slower proliferation rate, underwent cell cycle arrest in S phase, and displayed a higher percentage of apoptotic cells. Moreover, microarray analyses showed a change in the expression of signalling-related genes in these cells, suggesting that the cell proliferation defects in DDX51siRNA H1299 cells might be linked to a change in transcriptional regulation. DDX51siRNA H1299 xenografts in mice formed smaller tumours compared with control cells, suggesting that the protein also has a role in vivo. Taken together, these data suggest that DDX51 regulates cell proliferation by facilitating the transition between the S and G2 phases of the cell cycle, likely by affecting the transcriptome of different signalling pathways. These results also suggest that DDX51 plays an important role clinically, indicating that it is a potential target for future therapeutic applications.
Cancer patient samples. Material  Real-time quantitative PCR (RT-qPCR). RNA extraction was performed using TRIzol (Invitrogen, USA) according to the manufacturer's instructions. cDNA was synthetized by reverse transcription using an ABI 2720 thermal cycler (ABI Biosystems, USA) according to the manufacturer's instructions (M-MLV-RTase, Promega, USA). The cDNA product was detected using a SYBR Green Supermix kit (Toyobo, Osaka, Japan) with a Takara Bio PCR Thermal Cycler Dice Real Time TP800 (Takara, Japan). The cycling parameters were 95 °C for a 30-s hot start followed by 45 cycles of 95 °C for 5 s and 60 °C for 30 s. The relative mRNA expression (EIF3B/GAPDH) was determined using the 2 −ΔCt method. The sequences of the primers used are presented in Supporting Table SX. Lung cancer tissue microarrays. Lung cancer tissue microarrays (TMAs) were constructed at Shanghai Outdo Biotech Co., Ltd (Shanghai, China). Rabbit polyclonal anti-human DDX51 antibodies (1:150) were used for immunohistochemistry according to a two-step protocol. The protein expression patterns of DDX51 were analysed in 75 lung cancer tissues and paired adjacent noncancerous tissues. The staining intensity score was assigned as follows: 0 points, negative; 1 point, low; 2 points, medium; and 3 points, high. The staining percentage score was as follows: 0% positive cells, 0; 1-25% positive cells, 1; 26-50% positive cells, 2; 51-75% positive cells, 3; and ≥ 76% positive cells, 4. The "DDX51 final score" was calculated as the combination of the "staining intensity score" and the "staining percentage score". Patients were grouped into "low expression" (total score of 0-5) and "high expression" (total score of 6-12) groups according to their :DDX51 scores". The difference in DDX51 expression between tumour and non-tumour tissue was determined using paired Wilcoxon tests. The association between clinicopathological parameters and DDX51 expression was analysed using chi-square tests. The association between DDX51 expression and overall survival was evaluated using Mantel-Cox log-rank tests. Overall survival time was defined as the number of months from the date of histological diagnosis to the date of last contact or death from any cause. Patients who were alive at the last follow-up or were lost to follow-up were censored. Survival curves were plotted using the Kaplan-Meier method.
Cell proliferation assays. After achieving the logarithmic growth phase H1299 cells were trypsin-digested, resuspended in standard medium, and then seeded into 96-well plates at a density of 2,000 cells/well. The number of GFP fluorescence-positive cells was counted using a Cellomics Array Scan High Contents Screening Reader on five consecutive days.
Apoptosis assays. Apoptosis was assessed using annexin V-based flow cytometry using standard laboratory methodology. Briefly, cells were transfected as described above, and incubated for 5 days. The cells were then harvested, resuspended in binding buffer at a density of 1 × 10 6 cells/ml, and 100 μ l of this suspension was added to FACS tubes and stained with annexin V. Cells were mixed gently in a dark room for 15 min at room temperature, and then analysed using flow cytometry. Although standard flow cytometry apoptosis assays use co-staining with propidium iodide, we analysed only annexin V since the cells were transfected with a GFP-expressing vector and the excitation wavelengths of propidium iodide and GFP overlap.
Scientific RepoRts | 6:26108 | DOI: 10.1038/srep26108 Colony formation assays. After reaching the logarithmic growth phase H1299 cells were trypsinized, counted, and seeded at a density of 800 cells/well into six-well plates containing regular culture medium. After 14 days the cells were washed twice with PBS, fixed with methanol, and stained with Giemsa. The colonies were then photographed and scored.
Cell cycle assay. Lentivirus-transfected cells were cultured in 6-cm dishes until they reached 80% confluence, and were then trypsinized, washed twice in PBS, and fixed with 70% pre-chilled ethanol at 4 °C for 1 h. The fixed cells were washed and stained with a propidium iodide (PI) mixture containing 50 μ g/ml PI and 100 μ g/ml ribonuclease in PBS for 45 min at 37 °C. The cells were passed through a 300-mesh nylon net, before the DNA content was determined using quantitative flow cytometry with the standard optics of a FACScan flow cytometer (Becton-Dickinson FACS Calibur). All experiments were performed in triplicate.
Tumour growth in nude mice. H1299 cells (5 × 10 6 ) were suspended in 100 μ L serum-free Dulbecco's modified Eagle medium and Matrigel (BD Biosciences, San Jose, CA; 1:1) and implanted subcutaneously into   . T-tests were used to determine significance, with a p-value of < 0.05 to indicate a significant difference.

Results
In order to fully understand the role of DDX51 in NSCLS we used a siRNA approach to target the transcription of DDX51. We first assessed the effectiveness of our siRNA in the human NSCLC cell line H1299 ( Supplementary  Fig. 1), and then measured the proliferation rate of DDX51-siRNA lines. The Cellomics assay revealed that DDX51siRNA cells had a significantly slower proliferation rate than control cells after 3, 4, and 5 days (P < 0.01; Fig. 1A). We investigated this slow growth phenotype further by analysing the FACS profile of DDX51-siRNA cell cultures. The results indicated that there were a higher percentage of DDX51-siRNA H1299 cells in S phase compared with the control cells (P < 0.01). Conversely, there were fewer DDX51-siRNA than control H1299 cells in G2/M2 phase (P < 0.01; Fig. 1B). Overall, these data suggest that the DDX51 siRNA construct inhibited cell proliferation by blocking cell cycle progression in S phase. Because cell cycle arrest coincided with apoptosis, we next investigated whether the expression of DDX51 siRNA induced apoptosis in H1299 cells. FACS analysis of Annexin V-stained cells demonstrated that the percentage of DDX51-siRNA H1299 cells undergoing apoptosis was significantly higher (P < 0.01) compared with the control group (Fig. 1C).
Finally, colony formation assays demonstrated that the expression of DDX51 siRNA dramatically reduced the ability of H1299 cells to form colonies (Fig. 2), consistent with our cell proliferation and apoptosis observations. Because DDX51 plays a role in regulating RNA processing (10), it is possible that its effects on the cell cycle and apoptosis could be elicited by changes in the transcriptome of H1299 cells. Therefore, we profiled the gene expression of DDX51-siRNA H1299 cells using microarrays. Our analysis showed that H1299 DDX51-siRNA cells had significant changes in the transcription of several signalling-related genes compared with negative control cells (Supplementary Fig. 2). The results demonstrated that 122 and 137 genes were up-and down-regulated in DDX51 knockout cells, respectively. Genes upregulated significantly included RAPA1, MAP4K4, IL1R1, JUN, FOS, TGFBR2, and HSPA8. The genes downregulated significantly in the presence of reduced levels of DDX51 included MAP2K5 and MKNK2 (Supplementary Table 1). The microarray results were also assessed using GO analyses, and the results are shown in Supplementary Fig. 3 and Supplementary Table 2.
The microarray results were validated by analysing the expression of a group of representative proteins using western blotting. Data revealed that the expression of TGF-β R1, IL1-R1, and C-FOS was increased in cells expressing DDX51 siRNA (Fig. 3), confirming the microarray data. These data suggest that the effects of DDX51 siRNA on the cell cycle and apoptosis were likely dependent on the broader regulation of cellular functions.
To confirm the in vitro results we next investigated the effects of suppressing DDX51 expression in an animal model. Wild type and DDX51 siRNA-expressing H1299 cells were implanted into nude BALB/c nu/nu mice, and tumour progression was monitored for 1 week before the animals were sacrificed. The volume of DDX51-siRNA H1299 tumours was significantly smaller than control tumours throughout the experiment (Fig. 4A,B). Similarly, Figure 3. The effects of DDX51 knockdown on protein expression according to western blotting. GAPDH was used as a loading control. NC, negative control; KD, DDX51 knockdown. All results were reproducible in three independent experiments. *P < 0.05.
Scientific RepoRts | 6:26108 | DOI: 10.1038/srep26108 DDX51-siRNA H1299 tumours were lighter at the end of the observation period compared with control tumours (Fig. 4C). These results suggest that DDX51 plays a role in cancer growth in vivo.
To confirm this hypothesis and elucidate the mechanism of action of DDX51 in NSCLC in vivo we analysed tissues that had been harvested from hospitalized NSCLC patients (Table 1). DDX51 was overexpressed in cancer tissues compared with neighbouring regions (P < 0.05; paired sample test) according to both qRT-PCR and immunohistochemistry (Fig. 5A,B). Therefore, we investigated the relationship between DDX51 expression and clinicopathological factors in NSCLC patients (Table 1). There were no differences in any factors between patients with high or low levels of DDX51 expression (Table 1). Similarly, there was no difference in the survival of patients in the two groups (Fig. 5C). No other lung cancer risk factors, such as N or TNM stage, had a high correlation with DDX51 expression (Table 2).

Discussion
The current study demonstrated that DDX51 facilitates cell cycle progression and therefore promotes cell proliferation. Since the DDX51-siRNA cells were arrested in S-phase, we propose that DDX51 regulates the progression through S phase. The current results also suggest that this is the result of regulating a number of different signalling pathways, as indicated by the microarray data. This hypothesis is consistent with a role for DDX51 in ribosome assembly and RNA metabolism 10 , as well as with previous reports revealed a function for other DDX family members in regulating the cell cycle via different signalling pathways [11][12][13] .
The current microarray and western blot data demonstrated that DDX51 siRNA-expressing cells expressed higher levels of TGF-β -R, IL1R, and C-FOS. TGFβ has a dual role in lung cancer, since it both promotes and inhibits cell proliferation 16 . Some studies have suggested that the role of TGF-β might depend on its concentration. Although some studies have suggested that low concentrations were correlated with a role in tumour suppression 17,18 , knocking down DDX51 in the current study increased TGFBR2 levels, slowed cell proliferation, and delayed cancer progression. It is also difficult to explain the overexpression of C-FOS in DDX51-siRNA cells since  it is generally regarded as an oncogene 19 . Therefore, future studies are needed to address these apparently contradictory data. Although they were not studied functionally in the current study, other genes whose expression was altered in the DDX51 might also be important in NSCLC and warrant further investigation in future studies. For example, the expression of three MAPK-related genes was altered (MAP4K4, MAP2K5, and MKNK2), and this pathway plays an important role in the survival of a number of different tumour cells 20,21 . In addition, HSPA8 polymorphisms have been reported in a number of different types of cancer 22,23 .
The results of xenografts experiments confirmed that depleting DDX51 expression slowed cancer progression in an animal model. Although the current experiment did not analyse the effects of silencing DDX51 on mouse mortality, the reduced sized tumours formed by DDX51-silenced cells suggests that DDX51 might be a good target for cancer therapy. Consistent with this, our data suggested that DDX51 expression was significantly higher in cancer tissues compared with non-tumour control tissues. However, it is currently unclear if this is an indication of the severity of the cancer itself. While the current data indicate that inhibiting DDX51 expression reduced the proliferation of tumours, future studies will have to address whether the dysregulated expression of DDX51 alone is sufficient to induce uncontrolled cell replication. Because DDX51 expression correlated with cell proliferation and hence the cell cycle, these data are also consistent with the known role of DDX51 in ribosome processing 10 , as well as potentially protein translation during replication.
In conclusion, the current results confirmed the importance of the DDX family and pre-rRNA processing in the proliferation of cancer cells, and revealed a role for DDX51 in this process for the first time. Moreover, these data provide an important starting point for future studies investigating the potential of DDX51 as a target for pharmacological screens for the development of new potential anti-cancer drugs.
How to cite this article: Wang, X. et al. The DEAD-box RNA helicase 51 controls non-small cell lung cancer proliferation by regulating cell cycle progression via multiple pathways. Sci. Rep. 6, 26108; doi: 10.1038/ srep26108 (2016). This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/