Background
Cervical cancer (CeCa) is the fourth most common cancer in women worldwide, with more than 500,000 new cases each year and an estimated 260,000 deaths, of which approximately 87% occur in developing countries [
1]. Despite the implementation of vaccination against Human Papillomavirus infection, it will take decades before the impact on the cervical cancer incidence became clear, giving the natural history of this disease [
2]. Therefore, it is necessary to better understand the biology of this type of cancer in order to improve the early detection and treatment.
Voltage-gated sodium (Na
V) channels are heteromeric transmembrane proteins that consist of a single pore forming α-subunit associated with one or more auxiliary β-subunits (Na
Vβs) [
3]. These channels are known to be responsible for action potential generation and propagation in excitable cells, but they are also widely expressed and upregulated in a variety of human cancer types, including breast, prostate, colon, ovarian, lung, gastric, melanoma, astrocytoma and uterine cervix, where the Na
V α-subunit activity has been associated mainly, but not exclusively, with cell motility and invasiveness [
4‐
14]. So far, four Na
Vβs (Na
Vβ1–Na
Vβ4) have been described. These are multifunctional type I transmembrane proteins that modulate Na
V α-subunit gating, localization and transit to the plasma membrane. Additionally, they possess a V-type immunoglobulin repeat in the extracellular domain similar to the family of neural cell adhesion molecules (CAMs), thus, their role as CAMs in a trans-homophilic and trans-heterophilic way, and with other molecules, in presence or absence of the α-subunits has been described in numerous reports and in different expression systems [
15‐
22].
The role of Na
Vβs in cancer has also been studied, although not as widely as for the α-subunits. In prostate cancer (PCa), it has been found that the overexpression of Na
Vβs is associated with the increasing metastatic potential of PCa cell lines [
23], and when overexpressing Na
Vβ2 in LnCaP cells, the cells acquired a fibroblastic-like morphology, became more invasive and had enhanced migration in vitro; in contrast, these cells had a reduced tumor volume in vivo [
24,
25]. On the other hand, the overexpression of Na
Vβ3 in Saos-2 (osteosarcoma) and T98G (glioblastoma) cells, seems to activate an apoptotic pathway mediated by the tumor suppressor protein p53 [
26]. However, most of the studies have been done in breast cancer, where it has been demonstrated that Na
Vβ1 is overexpressed in the low metastatic MCF-7 cell line vs highly metastatic MDA-MB-231 cells [
27]. Overexpressing Na
Vβ1 in these last cells produced a drastic reduction on in vitro transwell migration and an increase in cellular adhesiveness. In contrast, MDA-MB-231 cells stably transfected with Na
Vβ1 produced larger tumors in an in vivo model compared with those generated by wild type MDA-MB-231 cells. In addition, overexpression of Na
Vβ1 induced a reduction in caspase activity in these cells [
28]. More recently, the potential role of Na
Vβ4 as a metastasis-suppressor gene has been described in breast cancer cells, associated with the small GTPase RhoA activity and therefore with the cytoskeleton remodeling during migration and invasiveness [
29].
We have previously demonstrated the overexpression of Na
V α-subunits in cervical cancer biopsies vs non-cancerous cervix, and furthermore, we reported the functional expression of these proteins in primary cultures derived from human cervical cancer biopsies and their contribution to the cell migration and invasiveness. Specifically, Na
V1.6 showed to be the most overexpressed α-subunit and the one responsible of the augmented invasive potential as the specific blockade of this channel was able to diminish the invasive potential with the same effectiveness as the blockade of all the Na
V α-subunits by TTX [
14,
30]. In the present work, we investigated the electrophysiological activity of Na
V channels and the expression of Na
Vβs in three cervical cancer cell lines (HeLa, SiHa and CaSki) as well as their contribution to three metastatic cell behaviors: in vitro proliferation, transwell migration and invasiveness.
Methods
Cell lines and cell culture
Cervical cancer cell lines SiHa, CaSki (both HPV-16 positive) and HeLa (HPV-18 positive) were grown in DMEM supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin at 37 °C in a humidified 5% CO2 incubator. All cell culture reagents were purchased from Gibco-Thermo Fisher Scientific (Waltham, MA). The three cell lines were authenticated by short tandem repeat profiling prior to the experiments (data not shown).
Overexpression and downregulation of NaVβs expression
Plasmids containing each of the Na
Vβs from rat were kindly donated by Dr. L. Isom (University of Michigan). Overexpression of each Na
Vβ in CeCa cells was achieved by transient transfections using the JetPEI transfection reagent (PolyPlus Transfection; Illkirch, France), according to the manufacturer’s instructions. On the other hand, downregulation of Na
Vβs expression was performed using two or three different predesigned siRNAs (Sigma-Aldrich; St. Louis, MO) for each Na
Vβ (100 nM each); in all cases, we chose the one with the best efficiency. Optimal siRNA transfection conditions were stablished previously by using 50 nM of siGLO Green Transfection Indicator (Sigma-Aldrich), a fluorescent siRNA for determining transfection efficiency. Efficiency of both experimental strategies to manipulate the expression levels of each Na
Vβ was evaluated by RT-PCR and western blot (see Additional file
1).
Electrophysiology
Detailed methods for whole-cell patch-clamp recordings and protocols have been previously described [
14,
30]. Sodium currents of native or transiently transfected CeCa cells were recorded at room temperature (20–23 °C) using an Axopatch 200B amplifier, a Digidata 1320A/D converter, and the pCLAMP 10.0 software (Molecular Devices; Sunnyvale, CA). As a positive control for sodium currents, we used transiently transfected CeCa cells with the Na
V1.6 channel.
Standard PCR (RT-PCR)
The Na
Vβs mRNA expression levels of CeCa cell lines were assessed by RT-PCR with specific primers for each β-subunit and β-actin as control (Table
1) as described before [
30]. Briefly, total RNA of each cell line was isolated with Trizol reagent (Invitrogen; Carlsbad, CA) and RT-PCR reactions were performed in a final volume of 25 µl using the Super-Script One-step RT-PCR kit (Invitrogen) with 250 ng of total RNA, 0.5 µl of enzyme mix, 0.2–0.5 µM of each primer, and 12.5 µl of 2X buffer containing 0.4 mM of each dNTP and 2.4 mM MgSO
4.
Table 1
RT-PCR primers information
NaVβ1 | NM_001037 | F: AGAAGGGCACTGAGGAGTTT R: GCAGCGATCTTCTTGTAGCA | 379 | 60 |
NaVβ2 | NM_004588 | F: GCCCACCCGACTAACATCTC R: ATGCGGAACTGGAGGAACA | 285 | 62 |
NaVβ3 | NM_018400 | F: CTGGCTTCTCTCGTGCTTAT R: TCAAACTCCCGGGACACATT | 353 | 60 |
NaVβ4 | NM_174934 | F: CACGCCACCATCTTCCTCCAA R: TGCAGCTGCTCAGCCCGAAG | 284 | 65 |
β-actin | NM_001101 | F: GCTCGTCGTCGACAACGGCTC R: CAAACATGATCTGGGTCATCTTCTC | 353 | 60 |
Real-time PCR (qPCR)
Total RNA was extracted using the RNeasy Mini Kit (Qiagen; Hilden, Germany), then reverse-transcribed with the High Capacity cDNA Reverse Transcription kit (Applied Biosystems; Foster City, CA) according to the manufacturer’s instructions using 2 µg of total RNA in a final volume of 20 µl. Real-time PCR was carried out in a Rotor-Gene Q (Qiagen) using Custom TaqMan Gene Expression Assays (Applied Biosystems) as described before [
14]. Briefly, 100 ng of cDNA, 0.4 µl of the TaqMan assay (Table
2) and 5 µl of TaqMan Universal PCR Master Mix (Applied Biosystems) were mixed in a final reaction volume of 10 µl for each qPCR reaction. At least three independent experiments were done, and each assay was performed in triplicate. The results were analyzed by the 2
−ΔΔCt method [
31] using HPRT1 expression as the normalizing gene control and results are shown as relative expression values of Na
Vβ1 in HeLa cells.
Table 2
qPCR primers information
NaVβ1 | NM_001037 | F: GGAGGATGAGCGCTTCGA R: CAGATCCTGCAGGTCTTTGGT P: CCCCGGCTGCCATT | 70 |
NaVβ2 | NM_004588 | F: TGCAGCCGGAGGATGAG R: GAGGACCTGCAGATGGATCTTG P: CCCCTGACCGCCACCG | 92 |
NaVβ3 | NM_018400 | F: CGCCAGCCCCAGAAGAT R: CACAGGGAAGCAGACACTGA P: TTTCCCCTGGCTTCTC | 90 |
NaVβ4 | NM_174934 | F: AAGAAGTGGACAACACAGTGACA R: TGAGTTTCTTGATCAGCAGGATGAG P: ACCCCGCCCACGACAG | 93 |
Western blot
Total protein from native or transiently transfected CeCa cells was extracted 24, 48, 72 and 96 h post-transfection (with cDNA or siRNAs, for overexpression or inhibition of the NaVβ expression respectively) using RIPA buffer (25 mM Tris–HCl, pH 7.4; 150 mM NaCl; 1% IGEPAL; 1% Sodium deoxycholate, and 1% SDS) supplemented with complete EDTA-free protease inhibitors (Roche, Switzerland), and quantified by Bradford assay. Equal amounts of protein (100 µg) were subjected to SDS-PAGE, transferred onto a polyvinylidene difluoride membrane (Millipore, Billerica MA) and probed overnight with the following primary antibodies: rabbit anti-NaVβ1 (1:3000; LifeSpan BioSciences Inc.; Seattle, WA); rabbit anti-NaVβ2 (1:1000; LifeSpan BioSciences Inc.); rabbit anti-NaVβ3 (1:5000; Abcam; Cambridge, UK), rabbit anti-NaVβ4 (1:3000; Novus Biologicals; Littleton, CO) and a homemade mouse anti-β-actin antibody (1:1000) used as a loading control. Blots were subsequently probed with an anti-rabbit or an anti-mouse (as the case may be) secondary antibody conjugated with horseradish peroxidase (1:10,000; Santa Cruz Biotechnology; Dallas, TX) for 1 h at room temperature and visualized using the SuperSignal West Pico chemiluminescent substrate (Thermo Fisher Scientific). Signal intensity of immunoblots was calculated with ImageJ software.
Proliferation assays
Cell proliferation was determined using the colorimetric 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay. Briefly, cells were plated in 48-well microplates at an initial density of 2 × 103 cells per well. To measure cell proliferation at 24, 48, 72 and 96 h after seeding, 10 µl of MTT solution (5 mg/ml) were added in each well after replacing the culture medium with 200 µl of fresh culture medium. The cells were incubated at 37 °C, and 3 h later, cell culture and MTT solution were removed and 150 µl of dimethyl sulfoxide was added into the plate to dissolve formazan crystals. The absorbance at a wavelength of 562 nm was recorded using an Eppendorf AG 22331 Biophotometer (Eppendorf; Hamburg, Germany). For analyzing the role of NaVβs on proliferation, after overnight incubation, cells were transfected with the respective cDNA to overexpress or the specific siRNA to downregulate each β-subunit expression, and finally, MTT assay was performed 24, 48, 72 and 96 h after transfection.
Cell cycle analysis by flow cytometry
The role of NaVβ1 on cell cycle in SiHa cells was assessed by flow cytometry after staining the cells with propidium iodide (PI; Molecular Probes, Life Technologies; Grand Island, NY). Briefly, 2 × 105 cells were seeded onto 12-well plates, incubated in normal conditions for 24 h, synchronized by serum-reduced conditions (0.2% FBS) for 48 h and then transfected with the specific cDNA or siRNA. After incubating for 48 h, cells were collected and fixed with ice-cold 70% ethanol, stored for at least 24 h at − 20 °C, treated with RNase and stained with PI for 1 h at 37 °C. For each condition, 2 × 104 cells (singlets) were analyzed in an Accuri C6 cytometer (BD Biosciences; San Jose, CA), and the percentages of cells in each cell cycle phase were analyzed with FlowJo software (Tree Star; Ashland, OR) using the Dean-Jett-Fox algorithm by three independent tests for each condition. Appropriate gating was used to select the single cell population and used on all samples.
In vitro migration and invasion assays
Migration and invasion assays were done using culture inserts with 8-µm pore size membranes covered or not with Matrigel (Corning Inc., Corning, NY) for invasion and migration respectively. The upper compartment was seeded with 5 × 104 for migration and 1 × 105 cells for invasion in DMEM supplemented with 5% FBS, and the lower compartment was filled with DMEM supplemented with 15% FBS as a chemoattractant. The number of cells migrating or invading (according to the circumstances) over 48 h at 37 °C was evaluated using the MTT assay described before. Results were compiled as the mean of at least three repeats, each time done by duplicate, and compared with control conditions.
In silico immunohistochemistry (IHC) staining analysis
Expression of Na
Vβ4 in CeCa and normal cervical tissues was examined using the Human Protein Atlas (
http://www.proteinatlas.org/) [
32,
33], a Swedish-based online tool for human protein expression analysis. Representative CeCa and normal tissues cores were chosen for illustrative purposes.
Statistical analysis
All quantitative results are given as the mean ± S.D. Differences in means were tested with an unpaired two-tailed Student’s t test.
Discussion
For many years, Na
Vβs were considered only as “auxiliary” Na
V subunits commissioned of regulating the α-subunit kinetics and cellular localization. Nowadays plenty of evidence shows the multifunctionality of these proteins, participating in a wide variety of cellular processes like cell adhesion, cytoskeleton remodeling, filopodia and invadopodia formation, transcriptional regulation, cell cycle regulation and even apoptosis [
21,
29].
For the present work, we chose three different CeCa cell lines in order to cover a wide range of CeCa cases, and to analyze the potential role of Na
Vβs in the malignant behavior of this disease. SiHa and CaSki cells are HPV-16 positive (the most frequently detected viral type in CeCa, found in approximately 57% of cases) whereas HeLa cells arise from a HPV-18 positive tumor (the second most frequent virus; 16% of cases) [
37]. Furthermore, HeLa and SiHa cells derived from a primary tumor, while CaSki cells were taken from a metastatic tumor. In addition, the patients’ ethnicity and the cellular origins of the three cell lines are different [
38‐
40].
Here we demonstrated the presence (expression) of Na
Vβs in the three different CeCa cell lines (Fig.
1), even in the absence of functional Na
V α-subunit expression in the plasma membrane (evidenced by whole-cell patch-clamp recordings, see Additional file
2), suggesting that their role in cancer cell biology is independent of the pore-forming Na
V α-subunit.
When we evaluated the aggressiveness of each cell line, by analyzing the basal proliferation, migration and invasiveness in vitro, HeLa cells demonstrated to be the most aggressive cell line, as these cells proliferated, migrated and invaded the most, followed by SiHa cells. CaSki cells were the less proliferative, and they migrated and invaded the less (Fig.
2). As far as we are concerned, this is the first report about the differences in the metastatic behavior among these CeCa cell lines.
We have previously reported the role of Na
V1.6 (the most overexpressed Na
V α-subunit in CeCa biopsies) in the invasiveness potential of CeCa primary cultures [
14]. In agreement with our recent findings [
34], in the present work we did not find voltage-gated sodium currents in any of the CeCa cell lines studied with the patch-clamp technique. This could be due to at least two possibilities: either the sodium channels are being correctly synthetized but are not reaching the plasma membrane for some unknown reason; or these cells are expressing a non-functional splicing variant. In this regard, it has been reported that Na
Vβs participate in the appropriate traffic and localization of the Na
V α-subunits in the plasma membrane [
15,
35,
36]. Moreover, it has been shown that Na
Vβ2 can be cleaved by secretases, generating small intracellular peptides capable of reaching the cell nucleus and promoting the transcriptional upregulation of genes including SCN1A, which encodes for Na
V1.1 [
41,
42]. Hence, we thought that maybe by modifying the expression of these proteins we could promote the functional localization of conducting Na
V channels in the plasma membrane, however our results suggest this is not the case, as the manipulation of the expression of the Na
Vβs in HeLa and SiHa cells was not enough to promote the appearance of voltage-dependent sodium currents (see Additional file
2). Yet, there are many other molecules such as hormones and growth factors that could be involved in the correct trafficking of the channel. In fact, we have examined the possibility that Na
Vs functional expression in CeCa cell lines could be regulated by β-estradiol or EGF, but we did not find sodium currents under any of the experimental conditions tested (data not shown). Another likely explanation could be related to the oxygen conditions in which the cell lines are cultured. It has been reported that hypoxia can change both the expression and activity of Na
Vs. Tumor cells usually experience severe hypoxia, which has been correlated with a more robust expression and activity of Na
Vs [
43,
44]. On the contrary, CeCa cell lines have been cultured in normoxic tissue culture conditions, which might have modified the Na
Vs expression and/or trafficking.
It is also possible that CeCa cell lines are expressing a non-conducting channel with a different and unknown function and/or localization in the cell. It remains to be investigated whether Na
V α-subunits are expressed in intracellular compartments, as well as their possible role in invasion. The existence of functional intracellular sodium channels has been demonstrated in macrophages where the activity of this sodium channel contributes to cellular invasion through a mechanism involving sodium ions release from cationic intracellular stores, followed by a subsequent mitochondrial calcium release mediated by the Na
+/Ca
2+ exchanger, which in turn facilitates cytoskeletal remodeling and invadopodia formation [
12]. In fact, recent data from our group suggests the presence of intracellular Na
V1.6 channels in CeCa cell lines [
34]. However, more experiments are needed to fully elucidate this discrepancy between Na
Vs functional expression in CeCa primary cultures vs cell lines.
Regarding to the effect of Na
Vβs on cell proliferation, we found that only Na
Vβ1 had an effect on this process and only in SiHa cells: the overexpression of Na
Vβ1 increased cell proliferation, whereas the downregulation of the expression of this protein decreased it (Fig.
3a). This is in agreement with observations reported by Nelson et al. [
28] demonstrating that Na
Vβ1 enhances breast tumor growth and metastasis in vivo, by increasing cell proliferation and reducing apoptosis. The fact that we only observed this effect in one out of three CeCa cell lines shows that it is a particularity of the cell line (SiHa) rather than a general role of Na
Vβ1 on proliferation in cervical cancer.
On the other hand, it has been suggested that Na
Vβ3 mediates a p53-dependent apoptotic pathway [
26]. In breast cancer cells, for example, this subunit is totally absent, although it is normally expressed in non-cancerous breast cells (Sanchez-Sandoval et al., unpublished data). However, the overexpression or down regulation of this subunit did not alter the proliferation rate of the CeCa cell lines analyzed here, suggesting a lack of the pro-apoptotic activity of Na
Vβ3 in these cells. This might be related to the p53 protein status in CeCa cell lines. It has been reported that the association between the E6 protein (one of the two main HPV oncogenes) with p53 leads to the specific ubiquitination and degradation of p53 protein [
45], therefore inactivating any pro-apoptotic effect due to the Na
Vβ3 expression in basal conditions. This is one of the several ways the HPV genome secures the growing of the tumor. However, the intermediate steps that Na
Vβ3 must undergo to promote apoptosis is still unknown, as well as why Na
Vβ3 had no pro-apoptotic effect in CeCa cells when overexpressed. A possible explanation could be the existence of a positive (and necessary) feedback regulation between p53 (or one of its multiple target genes) and Na
Vβ3, leading to a lack of pro-apoptotic activity of Na
Vβ3 in the absence of functional p53.
On the contrary, our migration experiments show a clear participation (negative regulation) of Na
Vβ1 on the ability of CeCa cells to perform transwell migration across of an 8 µm-pore membrane. When we induced the overexpression of this subunit, the migration rate decreased in the three CeCa cell lines, being HeLa cells the most affected (almost 50% inhibition in cell migration). Correspondingly, the inhibition of Na
Vβ1 expression by siRNAs led to significant increments in cell migration of HeLa, SiHa and CaSki cells (Fig.
4a). Thus, Na
Vβ1 could be acting as a cell adhesion molecule, promoting the cell–cell adhesion and therefore making it difficult to move across the membrane. It is also worth of notice that, among the three cell lines, HeLa have the lowest Na
Vβ1 expression compared to the other two CeCa cell lines (Fig.
1), and they also have the greatest migration capacity. In the same sense, CaSki cells are the ones with the bigger expression of Na
Vβ1, and the lowest migration behavior (Fig.
2b).
A study made on breast cancer cell lines revealed that the overexpression of Na
Vβ1 in highly metastatic MDA-MB-231 cells increased cell–cell adhesion and decreased in vitro migration, consistent with the proposed role of Na
Vβ1 as a cell adhesion molecule [
27]. However, the
trans-homophilic interactions through Na
Vβ1 can also mediate the process outgrowth in vivo, as in neurons, generating an elongated morphology and therefore potentiating metastasis, as has been reported for the same breast cancer cells but using mouse models instead of transwell experiments [
28]. Something similar could be happening with CeCa cell lines. Thus, the persistent contribution of Na
Vβ1 in the three CeCa cell lines migration is indicative of its role as a negative regulator of the in vitro migration, regardless the HPV type present in the CeCa cell line. Although it remains to be investigated whether this behavior also occurs in vivo.
With respect to the contribution of Na
Vβs in invasiveness, our results show that downregulation of Na
Vβ4 expression resulted in a significant increase in the in vitro invasiveness of the three cell lines studied here (Fig.
4b), suggesting the role of this subunit as a metastasis suppressor protein. These observations agree with the in silico analysis of Na
Vβ4 expression performed with the Human Protein Atlas data: normal cervix biopsies show moderated Na
Vβ4 expression, while in CeCa biopsies it seems to be absent (see Additional file
5). Taken together, these results strongly suggest that Na
Vβ4 expression is downregulated in CeCa, which in turn increases the invasive capacity of the cells. In accordance with this interpretation, out of the three CeCa cell lines studied in the present work, the HeLa cell line showed the lowest expression of Na
Vβ4 and the highest index of cell invasion.
In breast cancer cell lines, it has been found that increasing Na
Vβ1 expression promotes the overexpression of Na
V1.5, the main Na
V α-subunit involved in the invasiveness of this type of cancer [
27]. However, according to our results, this is not the case for CeCa cell lines, as the overexpression of Na
Vβ1 or Na
Vβ4 did not promote the expression of conducting Na
V1.6 channels in the plasma membrane, the principal Na
V α-subunit involved in the invasive potential of these cells. Thus, we conclude that the pathways used by these proteins to induce cell migration and invasion, are independent of the activity of Na
V1.6 or other Na
V α-subunits in the plasma membrane of CeCa cell lines.
Other groups have also reported the downregulation of Na
Vβ4 expression in cancer vs normal tissue. This is the case of prostate cancer [
23], papillary thyroid cancer [
46], and breast cancer [
29]. The latter study, performed by the group of Dr. Sebastien Roger at the University of Tours, reported solid evidence about the possible mechanistic pathway through which Na
Vβ4 prevents metastasis in cancer cells. They showed an increased activity of the small GTPase RhoA (crucial for cytoskeleton remodeling during migration and invasion) when Na
Vβ4 was silenced in breast cancer MDA-MB-231 cells. This effect was independent of Na
V1.5 channel activity, the main Na
V α-subunit implicated in BCa invasiveness. A similar mechanism might be taking place in CeCa cells: a downregulation of the Na
Vβ4 subunit that leads to an increased activity of RhoA and therefore, a more robust invasive potential, which is independent of the Na
V1.6 expression in the plasma membrane. However, we speculate that additional cellular mechanisms might be involved in this phenomenon, as the effects of Na
Vβ4 in cellular migration were not significant in our study, meaning that the sole change on cellular motility cannot explain the dramatic increase on cellular invasiveness when Na
Vβ4 is downregulated. A more recent study showed that Na
Vβ4 is downregulated in papillary thyroid cancer (PTC) compared with normal thyroid tissues. More interestingly, the authors also found that preserved Na
Vβ4 expression might independently predict favorable recurrence-free survival in classical PTC [
46]. Altogether, these data reinforce the notion of using Na
Vβ4 as a biomarker for cancer metastasis and a potential new therapeutic target for the treatment of cervical cancer.