Cucurbitacin E inhibits the proliferation of glioblastoma cells via FAK/AKT/GSK3β pathway

Introduction

Glioblastoma (GBM) is the most common malignant tumor of the brain, accounting for >50% of primary intracranial malignant tumors, and is also regarded as one of the most intractable early-death solid tumors in neurosurgery due to its strong aggression, rapid postoperative recurrence and high mortality (1). At present, GBM therapy continues to be a challenging medical issue. Multiple drug resistance and a high recurrence rate are the two main barriers to effective treatment; GBM therapy resistance has been attributed to tumor heterogeneity, hypermutation, hypoxia and immune-suppressive tumor microenvironment (2,3). Despite progress in the main treatment modalities for GBM, including surgery, radiotherapy and chemotherapy, the outcome for patients remains almost generally fatal, with a median survival of <2 years (4,5). Therefore, it is critically necessary to identify more effective therapeutic targets and improved therapeutic strategies for the treatment of GBM.

With the rapid advancement of natural medicines, the outstanding antitumor activity of traditional Chinese medicine monomers has steadily drawn more attention in recent years. Cucurbitacin E (CUE), a highly oxidized tetracyclic triterpene compound isolated from species of the genus Cucurbita, has been reported to exert anti-inflammatory and anti-analgesic properties (6). During the past few years, numerous studies have demonstrated that CUE has also anticancer effects, including inhibiting the proliferation of various cancer types, including gastric, liver, lung and colon cancers (7,8), as well as the capacity of inducing apoptosis and G2/M arrest in a number of cancer cells (9,10). CUE can also disrupt the cytoskeleton of actin and vimentin in prostate cancer cells, alter the morphology of tumor cells (11,12), and inhibit angiogenesis in human prostate tumors through the Janus kinase 2 (JAK2)-signal transducer and activator of transcription 3 (STAT3) signaling pathway mediated by vascular endothelial growth factor receptor 2 (VEGFR2) (13). Additionally, it has been revealed that CUE can inhibit Yes-associated protein signaling pathway and brain metastases of human non-small cell lung cancer (14), suppress the proliferation and invasion of osteosarcoma cells through phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway (15) and induce cellular senescence in colon cancer via modulating the miR-371b-5p/transcription factor AP-4 (TFAP4) axis (16). Of note, a previous study has shown that CUE inhibits the growth of GBM by arresting the cell cycle at G2/M phase through GADD45γ gene expression and blockade of cyclinB1/CDC2 complex (17). There is also evidence suggesting that CUE delays the onset of mitosis in GBM cells by upregulating GADD45β rather than downregulating the phosphorylation of extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) (18). However, the underlying mechanisms by which CUE exerts its anticancer effects on GBM remain to be fully elucidated.

Receptor tyrosine kinases (RTKs), a family of cell surface receptors, play critical roles in cell proliferation, survival and migration (19). The epidermal growth factor receptor (EGFR), which belongs to the ErbB family of RTKs, is frequently mutated and/or overexpressed in numerous human cancers (20). Once activated, EGFR activates numerous downstream signaling pathways, including mitogen-activated protein kinase (MAPK), PI3K/AKT, JAK/STAT and protein kinase C (PKC), which are highly associated with the proliferation, migration, angiogenesis and apoptosis of tumor cells (21–23). Focal adhesion kinase (FAK) plays an important role in tumor development. Previous studies have revealed that FAK activation promotes breast cancer angiogenesis (24). In an FAK-deficient mouse model, the cell motility is reduced and focal adhesion contact formation is enhanced (25). In addition, inhibition of the FAK and EGFR signaling pathways synergistically promotes apoptosis in breast cancer cells (26).

Protein kinase B (AKT), a serine/threonine kinase, is dysregulated in human cancers and plays a crucial role in tumor growth (27). Once activated, it transmits signals to numerous downstream effectors, including glycogen synthase kinase-3β (GSK3β) and forkhead box protein O (FOXO) (27). Collected evidence suggests that the FAK/AKT/GSK3β pathway is essential for the proliferation, invasion and metastasis of numerous cancers, including GBM (28–32). Furthermore, it has been identified that dysregulation of the cell cycle checkpoint proteins, including cyclinB1 and cyclinD1, contributes to the uncontrolled cellular growth and tumorigenesis (33). For example, cyclinB1 and cyclinD1 are overexpressed in numerous cancers which are involved in neogenesis and progression (34,35). However, it is not clear whether these signaling cascades are involved in the antitumor effect of CUE.

In the present study, it was aimed to demonstrate the effects of CUE on the proliferation of GBM cells and reveal the possible underlying molecular mechanisms.

Materials and methods

Antibodies and reagents

CUE was purchased from MedChemExpress. Antibodies for western blot analysis, including phospho-FAK (Tyr397) (1:2,000; cat. no. 8556S), FAK (1:2,000; cat. no. 3285S), phospho-AKT (Ser473) (1:2,000; cat. no. 4058S), AKT (1:2,000; cat. no. 4691S), phospho-GSK3β (1:2,000; cat. no. 9336S), GSK3β (1:2,000; cat. no. 9315S; all from Cell Signaling Technology, Inc.), cyclinB1 (1:2,000; cat. no. ab181593), cylinD1 (1:10,000; cat. no. ab134175), cyclinA2 (1:2,000; cat. no. ab181591), cyclinE1 (1:5,000; cat. no. ab133266; all from Abcam), Proliferating Cell Nuclear Antigen (PCNA; 1:2,000; cat. no. 10205-2-AP; Proteintech Group, Inc.), Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; monoclonal, rabbit anti-mouse; 1:5,000; cat. no. KC-5G4; Zhejiang Kangchen Biotech, Co., Ltd.), goat anti-rabbit IgG HRP-linked antibody (1:5,000; cat. no. 31460) and goat anti-mouse IgG HRP-linked antibody (1:5,000; cat. no. 31431; both from Thermo Fisher Scientific, Inc.). PF-562271 (FAK selective inhibitor, 10 µM) (cat. no. HY-10459) was purchased from MedChemExpress. EGF (cat. no. AF-100-15; 20 ng/ml) was purchased from PeproTech, Inc. Reagents used for cell culture including Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS) and trypsin-EDTA solution were purchased from Gibco; Thermo Fisher Scientific, Inc.

Cell culture

Human GBM cell lines U87-MG (derived from GBM of unknown origin; American Type Culture Collection no. HTB-14; cat. no. TCHu 138) and U251-MG (cat. no. TCHu 58) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) in 2018. Authentication testing of U87-MG and U251-MG cell lines was performed by Shanghai Biowing Applied Biotechnology Co. Ltd. via STR profiling. STR profiles match the standards recommended for U87-MG and U251-MG cell lines authentication. U87-MG and U251-MG cells were cultured in DMEM containing 4.5 g/l glucose supplemented with 10% FBS. Cells were maintained at 37°C in an incubator with 5% CO2.

Cell viability

A total of 2.0×103 cells/well were seeded onto 96-well plates and incubated at 37°C in an incubator with 5% CO2. After adherence, the cells were incubated with different concentration of CUE (0, 0.01, 0.025, 0.25, 2.5 and 25 µM) for 24 h. Then, 10 µl Cell Counting Kit-8 (CCK-8; cat. no. CK04; Dojindo Laboratories, Inc.) reagent was used to detect the cell viability, and the optical density (OD) at 450 nm was measured. Cell survival was expressed as fold of the control group. The optimal treatment concentration was 2.5 µM and thus was selected for the following experiments.

Western blot analysis

Cells were cultured overnight at 37°C in an incubator. After treatment, the cells were washed with PBS, and cell lysis buffer (cat. no. R0010; Beijing Solarbio Science & Technology Co., Ltd.) was added at 4°C for 25 min. The lysate was centrifuged at 12,000 × g at 4°C for 15 min. The protein concentration was measured by bicinchoninic acid protein assay. The equivalent amounts of protein (20 µg) were loaded into 10% SDS-PAGE gel (30% acrylamide) and separated at 110 V voltage. The proteins in the gel were transferred to a PVDF membrane and blocked in 5% skim milk for 2 h at room temperature. Subsequently, the membrane was incubated with different primary antibodies including phospho-FAK, FAK, phospho-AKT, AKT, phospho-GSK3β, GSK3β, cyclinB1, cylinD1, cyclinA2, cyclinE1 and GAPDH overnight at 4°C and then incubated with HRP-conjugated anti-rabbit IgG secondary antibody (1:5,000) or HRP-conjugated anti-mouse IgG secondary antibody (1:5,000) for 90 min at room temperature. Immunoreactive bands were visualized by enhanced chemiluminescence (Pierce; Thermo Fisher Scientific, Inc.) and semi-quantified using ImageJ software (version 1.47t; National Institutes of Health).

Immunofluorescence confocal microscopy

A total of 6.0×104 cells/well were seeded onto 12-well plates and treated with 2.5 µM CUE for 24 h. After washed with pre-cooled PBS, the cells were fixed with 4% paraformaldehyde for 15 min followed by permeabilization with 0.2% Triton X-100 in PBS for 10 min. Subsequently, the cells were blocked in PBST (0.1% Tween-20) containing 1% bovine serum albumin (cat. no. A8020; Beijing Solarbio Science & Technology Co., Ltd.) for 40 min. After washing with pre-cooled PBS, the cells were incubated with anti-human rabbit Ki67 antibody (1:1,000; cat. no. 27309-1-AP; Proteintech Group, Inc.) overnight at 4°C in a wet box. Next, the cells were then washed with pre-cooled PBS and incubated with anti-rabbit Alexa Fluor® 594 secondary antibody [(1:500; cat. no. Ab150080; Abcam), Excitation wavelength: 590 nm; Emission wavelength: 617 nm] for 1 h. The nucleus was stained with DAPI (cat. no. AR1176, no dilution; Boster Biological Technology Co., Ltd.), and the fluorescence signal was detected under an inverted fluorescent microscope.

Statistical analysis

Data are presented as the mean ± SEM of at least three independent experiments and were analysed via GraphPad Prism 7.0 (Dotmatics). Unpaired Student's t-test (two groups) or one/two-way ANOVA with Bonferroni's multiple comparison tests (more than two groups) were used for comparisons. P<0.05 was considered to indicate a statistically significant difference.

Results

CUE inhibits the proliferation of GBM cells

The effect of CUE on the growth of GBM cells was first examined using CCK-8 assay. Compared with the control group, CUE significantly inhibited the proliferation of U87-MG and U251-MG cells in a dose-dependent manner (Fig. 1A and B). The half maximal inhibitory concentration was ~2.5 µM. To further detect the influence of CUE on cell viability at different treatment times, the cells were exposed to 2.5 µM CUE for 24, 48 or 72 h, respectively. The results demonstrated that CUE significantly inhibited the proliferation of GBM cells in a time-dependent manner (Fig. 1C and D).

Figure 1. CUE inhibits the proliferation of glioblastoma cells. (A and B) Dose-dependent effects of CUE on the proliferation of glioblastoma cells. (A) U87-MG and (B) U251-MG cells were treated with increasing concentrations of CUE as indicated for 24 h and the cell viability was measured using Cell Counting Kit-8 assay. (C and D) Time-dependent effects of CUE on the proliferation of glioblastoma cells. (C) U87-MG and (D) U251-MG cells were treated with CUE (2.5 µM) for 24, 48 and 72 h. (E and G) The expression of PCNA in glioblastoma cells was measured by western blot after CUE treatment. (F and H) Quantitation of western blot results shown in panels E and G. (I and K) The expression of Ki67 in glioblastoma cells was measured by immunofluorescence after CUE treatment (Scale bar, 50 µm; 20X objective). (J and L) Quantitation of fluorescence level results demonstrated in panels I and K. Data are presented as the mean ± SEM. *P<0.05, **P<0.01 and ***P<0.001 vs. the control group. CUE, Cucurbitacin E.

To further confirm the effect of CUE on the proliferation of GBM cells, the expression of PCNA and Ki67, which are important indicators of tumor cells proliferation, was then detected. The cells were treated with 2.5 µM CUE, and then the expression of Ki67 and PCNA were measured by immunofluorescence and western blot assays, respectively. It was observed that the expression of PCNA was significantly downregulated in CUE-treated cells as compared with the control group (Fig. 1E-H). In addition, red fluorescence intensity of the experimental group was significantly decreased, indicating that CUE-treated cells had less ki67 expressed than untreated cells (Fig. 1I-L). Taken together, these results suggested that CUE significantly inhibits the proliferation of GBM cells (Fig. 1E-L).

CUE downregulates the phosphorylation level of FAK, AKT and GSK3β in GBM cells

FAK is a tyrosine kinase, which is closely related to the occurrence and development of tumors. Numerous studies have shown that downregulating the expression of FAK greatly inhibits the proliferation, migration and invasion of tumor cells (32,36,37). To evaluate whether the FAK-mediated signal transduction pathways, including AKT and GSK3β, were engaged in the antitumor effect of CUE, the phosphorylation level of FAK, AKT and GSK3β was measured in GBM cells following 2.5 µM CUE treatment at various time-points (0, 0.5, 1, 2, 8 and 24 h). The results revealed that CUE significantly decreased the phosphorylation level of FAK, AKT and GSK3β in U87-MG cells (Fig. 2A-D). Similar results were observed in U251-MG cells (Fig. 2E-H). These results suggested that CUE may inhibit the proliferation of GBM cells through FAK/AKT/GSK3β signaling pathway.

Figure 2. CUE downregulates the phosphorylation level of FAK, AKT and GSK3β in glioblastoma cells. (A and E) (A) U87-MG cells and (E) U251-MG were treated with CUE (2.5 µM) for the indicated time intervals. FAK, AKT and GSK3β phosphorylation were determined using western blot analysis. (B-D and F-H) Quantitation of the western blot results as shown in panels A and E. Data are presented as the mean ± SEM. *P<0.05, **P<0.01 and ***P<0.001 vs. the control group. CUE, Cucurbitacin E; FAK, focal adhesion kinase; p-, phosphorylated.

CUE blocks EGF-induced FAK, AKT and GSK3β phosphorylation in GBM cells

It has been reported that EGF upregulates the phosphorylation level of FAK, AKT and GSK3β in GBM cells (38–40). The effects of CUE on EGF-induced FAK, AKT and GSK3β phosphorylation were then explored. It was demonstrated that EGF (20 ng/ml) significantly increased the phosphorylation of FAK, AKT and GSK3β in both GBM cell lines, which were significantly blocked by CUE pre-treatment (Fig. 3A-D). In order to confirm whether AKT and GSK3β were in the downstream of FAK, the impact of PF-562271 (10 µM), a selective inhibitor of FAK, was examined on EGF-induced AKT and GSK3β phosphorylation in GBM cells. The results showed that PF-562271 significantly inhibited EGF-induced AKT and GSK3β phosphorylation (Fig. 3E-H), suggesting that AKT and GSK3β were downstream of FAK.

Figure 3. CUE blocks EGF-induced FAK, AKT and GSK3β phosphorylation in glioblastoma cells. (A and C) Effects of CUE on EGF-induced FAK, AKT and GSK3β phosphorylation in (A) U87-MG cells and (C) U251-MG cells. Cells were pre-treated with CUE (2.5 µM) for 24 h and then together treated with EGF (20 ng/ml) for 10 min. (B and D) Quantitation of western blot results as shown in panels A and C. (E and G) Effects of PF-562271 on EGF-induced FAK, AKT and GSK3β phosphorylation in (E) U87-MG cells and (G) U251-MG cells. Cells were pre-treated with PF-562271 (10 µM) for 30 min and then together treated with EGF for 10 min. (F and H) Quantitation of western blot results as shown in panels E and G. Data are presented as the mean ± SEM. *P<0.05, **P<0.01 and ***P<0.001 vs. the control group. CUE, Cucurbitacin E; FAK, focal adhesion kinase; p-, phosphorylated.

CUE inhibits the proliferation of GBM cells through the EGF-mediated FAK/AKT/GSK3β signaling pathway

After confirming that CUE inhibited the proliferation of GBM cells and blocked EGF-induced FAK/AKT/GSK3β phosphorylation, it was then investigated whether CUE inhibits the proliferation of GBM cells through the FAK/AKT/GSK3β signaling pathway. To this end, the effect of CUE on EGF-induced cell proliferation was first explored. It was found that CUE (2.5 µM) significantly inhibited the proliferation of EGF-induced GBM cells after treating the cells with it, independently of EGF (20 ng/ml) being present or absent (Fig. 4A and B). Similarly, FAK-specific inhibitor PF-562271 (10 µM) also significantly inhibited EGF-induced proliferation of GBM cells (Fig. 4C and D). Collectively, these results indicated that FAK/AKT/GSK3β signaling pathway is involved in the anti-proliferative effect of CUE in GBM cells.

Figure 4. CUE inhibits proliferation of glioblastoma cells through the EGF-mediated FAK/AKT/GSK3β signaling pathway. (A and B) Effects of CUE on EGF-induced cell proliferation in (A) U87-MG cells and (B) U251-MG cells. Cells were treated with CUE (2.5 µM) in the absence or presence of EGF (20 ng/ml) for 72 h and then cell viability was measured using CCK-8 assay. (C and D) Effects of PF-562271 on EGF-induced cell proliferation in (C) U87-MG cells and (D) U251-MG cells. Cells were treated with PF-562271 (10 µM) in the absence or presence of EGF (20 ng/ml) for 72 h and then cell viability was determined using CCK-8 assay. Data are presented as the mean ± SEM. ***P<0.001 vs. the control group. CUE, Cucurbitacin E; CCK-8, Cell Counting Kit-8.

CUE inhibits the expression of cyclinB1 and cyclinD1 in GBM cells

The effects of CUE on the expression of cyclinB1 and cyclinD1, two cyclins which play crucial roles in the proliferation of tumor cells, were also evaluated. In GBM cells treated with 2.5 µM CUE, cyclinB1 and cyclinD1 expression significantly decreased as compared with the control group (Fig. 5A-F). Notably, in contrast to cyclinB1 and cyclinD1, CUE had no discernible effect on cyclinA2 and cyclinE1 expression (Fig. S1).

Figure 5. CUE inhibits the expression of cyclinB1 and cyclinD1 in glioblastoma cells. (A) U87-MG and (B) U251-MG cells were treated with CUE (2.5 µM) for the indicated time intervals. Expression levels of cyclinB1 and cyclinD1 were measured using western blot analysis. (C-F) Quantitation of western blot results as shown in panels A and B. Data are presented as the mean ± SEM. *P<0.05, **P<0.01 and ***P<0.001 vs. the control group. CUE, Cucurbitacin E.

Discussion

In the present study, the effect of CUE on GBM cell proliferation and its underlying molecular mechanisms were investigated. The results demonstrated that CUE reduced the expression of cyclinB1 and cyclinD1, downregulated the FAK/AKT/GSK3β signaling pathway and consequently inhibited the proliferation of GBM cells.

Traditional Chinese medicine offers a number of drug candidates for cancer treatment. CUE, a member of the Cucurbitaceae family, inhibits the growth of multiple cancers (7,8). Previous studies have stated that CUE inhibited breast tumor metastasis (41) and induced autophagy of cancer cells by decreasing mTORC1 signaling and increasing Adenosine 5′-monophosphate (AMP)-activated protein kinase (AMPK) activity (42). In addition to its anticancer properties, CUE also improved liver fibrosis and inhibited the production of inflammatory factors (43,44). However, the effect of CUE on GBM is rarely reported except in two previous studies showing that CUE could induce mitosis delay in GBM cells by upregulating GADD45β and inhibit GBM growth via arresting the cell cycle at G2/M phase (17,18). In the present study, it was found that CUE significantly decreased the cell viability of GBM cell lines U87-MG and U251-MG in a dose-dependent and time-dependent manner. This finding was further confirmed by measuring the expression of Ki67 and PCNA, which are nuclear antigens linked to dividing cells, and both can be used to assess the level of cell proliferation (45). Using an immunofluorescence assay, it was identified that the amount and intensity of Ki67 fluorescence in cells treated with CUE were significantly reduced. Furthermore, it was revealed by western blot analysis that CUE suppressed PCNA expression. These findings indicated that CUE significantly inhibits the proliferation of U87-MG and U251-MG cells.

FAK, which is overexpressed and phosphorylated in various advanced solid tumors, is essential for tumor growth, proliferation and metastasis (46,47). Thus, FAK has become a potential target for cancer therapy (48). AKT is one of the best-characterized kinases known to regulate multiple cellular functions through phosphorylation of various substrates. Previous studies have demonstrated its critical role in the survival and death of cancer cells (49). As a downstream target of AKT, GSK3β signaling pathway is one of the crucial signal transduction pathways implicated in the development of numerous cancers (50). Although it has been established that AKT and GSK3β can be downstream signaling molecules of FAK, it is unknown if the FAK/AKT/GSK3β signaling pathway contributes to the GBM development process. In the present study, it was identified that CUE significantly downregulated the phosphorylation of FAK, AKT and GSK3β in GBM cells, but the exact molecular mechanism of how CUE affects the phosphorylation of FAK, AKT and GSK3β remains to be further studied. Notably, it was also found that CUE significantly blocked EGF-induced phosphorylation of FAK, AKT and GSK3β. Considering that numerous evidences have indicated that EGFR is frequently overexpressed in human cancers and overactivation of EGFR signaling cascades are highly associated with the occurrence and development of tumors (51,52), the aforementioned finding further confirmed the potential value of CUE in tumor therapy. Notably, in the present study, it was not explored which enzymes phosphorylate the FAK/AKT/GSK3β. It is worth noting that previous studies reported that FAK is downstream of EGFR. Once activated, EGFR transmits signals to the downstream Src/FAK pathway and the phosphorylation of the Src/FAK complex can initiate the activation of the MAPK or PI3K/AKT pathway (53). Furthermore, it has been reported that Src-3Δ4 mediates the interaction of EGFR with FAK and leads to EGF-induced FAK phosphorylation (54). Such evidence suggests that EGFR and Src-3Δ4 may be promising candidate enzymes that phosphorylate the FAK/AKT/GSK3β.

Another important finding in the present study was that CUE reduced the expression of cyclinB1 and cyclinD1 in GBM cells. Both cyclinB1 and cyclinD1 are important cell cycle-driven proteins. CyclinB1 regulates the G2/M phase transition of the cell cycle (33), while cyclinD1 regulates the G1/S phase transition of the cell cycle (55). Numerous studies have demonstrated that downregulation of cyclinB1 or cyclinD1 would lead to mitotic block and would inhibit the proliferation of numerous tumor cells (56–59). However, whether the reduced expression of cyclinB1 or cyclinD1 is related to the antiproliferation effect of CUE, and whether FAK/AKT/GSK3β pathway participated in this process remains to be further investigated.

While the current study revealed that CUE inhibited the proliferation of U87-MG and U251-MG cells by modulating the FAK/AKT/GSK3β signaling pathway, certain important questions remain to be answered. For example, in the present study, the antitumor effect of CUE on cultured cell lines was only evaluated; thus, it would be interesting to further validate the antitumor effect of CUE on the growth of GBM using tumor xenograft animal models, and then isolate the tumor tissues to detect the phosphorylation of FAK, AKT and GSK3β to further elucidate the molecular mechanisms. In addition, whether CUE has any effect on GBM migration, invasion and apoptosis, and whether the FAK/AKT/GSK3β signaling pathway is involved in these processes remains unknown. Answering these important questions will help us fully understand the antitumor effect of CUE.

In conclusion, the present study demonstrated that CUE exerts a distinct antitumor effect on GBM cells. CUE may inhibit the proliferation of GBM cells through the FAK/AKT/GSK3β signaling pathway. The present finding provides a promising basis for the development of effective new drugs for GBM therapy.

Supplementary Material

Supporting Data

Acknowledgements

Not applicable.

Funding

The present study was supported by the National Natural Science Foundation of China (grant no. 32160184), the Scientific Project of Jiangxi (grant no. 20181BAB215018) and the Department of Public Health of Jiangxi (grant no. 20185226).

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

PH and WC designed the study. WC conducted the majority of the experiments. FL, XL and LL contributed to the data collection and statistical analysis. WTC performed parts of the western blot experiments. TZ and YL performed immunofluorescence experiments. LN and YZ analysed the western blot data. WC and PH wrote the manuscript. All authors read and approved the final manuscript. PH and WC confirm the authenticity of all the raw data.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References