Silencing oncogene expression in cervical cancer stem-like cells inhibits their cell growth and self-renewal ability

Article metrics

Abstract

Accumulating evidence supports the concept that cancer stem cells (CSCs) are responsible for tumor initiation and maintenance. They are also considered as an attractive target for advanced cancer therapy. Using a sphere culture method that favors the growth of self-renewal cells, we have isolated sphere-forming cells (SFCs) from cervical cancer cell lines HeLa and SiHa. HeLa-SFCs were resistant to multiple chemotherapeutic drugs and were more tumorigenic, as evidenced by the growth of tumors following injection of immunodeficient mice with 1 × 104 cells, compared with 1 × 106 parental HeLa cells required to grow tumors of similar size in the same time frame. These cells showed an expression pattern of CD44high/CD24low that resembles the CSC surface biomarker of breast cancer. We further demonstrated that HeLa-SFCs expressed a higher level (6.9-fold) of the human papillomavirus oncogene E6, compared with that of parental HeLa cells. Gene silencing of E6 with a lentiviral-short-hairpin RNA (shRNA) profoundly inhibited HeLa-SFC sphere formation and cell growth. The inhibition of cell growth was even greater than that for sphere formation after E6 silence, suggesting that the loss of self-renewing ability may be more important. We then measured the expression of self-renewal genes, transformation growth factor-beta (TGF-β) and leukemia-inhibitory factor (LIF), in shRNA-transduced HeLa-SFCs and found that expression of all three TGF-β isoforms was significantly downregulated while LIF remained unchanged. Expression of the Ras gene (a downstream component of TGF-β) was also markedly decreased, suggesting that the growth-inhibitory effect could be via the TGF-β pathway. The above data indicate RNA interference-based therapy may offer a new approach for CSC-targeted cancer therapy.

Introduction

The concept of cancer stem cells (CSCs) is that tumors contain a small proportion of self-renewal and pluripotent cancer cells that are responsible for tumor initiation and maintenance.1 The high expression of multiple drug-resistant gene2 and ATP-binding cassette transporter3, 4 by these cells can also protect them from chemotherapeutic drugs. Therefore, CSCs are considered as an important target for developing future cancer therapies or improving the current therapies. However, as they are multi-drug resistant and also resistant to other current therapies (for example, radiotherapy),1 conventional methods will not be effective for targeting them and hence alternative approaches are required.

The RNA interference (RNAi) technique has been widely used as a tool for gene-functional studies and also has great potential for developing therapies against viral infection, genetic disorders and cancers.5, 6, 7, 8, 9 Several RNAi therapies have been used clinically, while many more are in clinical trials.10 For RNAi-based cancer therapy, oncogenes are obviously ideal targets, as they are the driving force of cancer cell growth and are highly expressed in cancer cells.11, 12 Although it is not clear whether oncogene expression in CSCs is different from that in other cancer cells, it is believed that these genes are vital for them. Therefore, oncogenes may offer a new approach for CSC-targeted cancer therapies. There are numerous studies on RNAi-based cancer therapy targeting oncogenes in various cancers; for example, human papillomavirus (HPV) E6 and E7 for cervical cancer,13, 14, 15, 16 and Her-2,17 c-myc18 and hdm-219 for breast cancer. However, there is no report on RNAi-based oncogene silencing in CSCs.

Cervical carcinoma is the second most common cancer in women worldwide and is highly associated (99%) with infections of high-risk types of HPVs. Viral early genes E6 and E7 from high-risk HPV types are responsible for the transformation of epithelial cells, and their continuous expression is essential for ongoing cervical cancer cell survival as they function as oncogenes.20, 21 Therefore, E6 and E7 are ideal targets for RNAi therapy. In recent years, there have been a number of publications showing the potential use of RNAi as a treatment for cervical cancer.14, 15, 16, 22, 23, 24, 25 We demonstrated that RNAi triggered by short-hairpin RNA (shRNA) targeting a specific site within E7 mRNA could even induce immunity to E7 in immune-competent mice.26 All the data indicate that RNAi-based therapy can be developed as a promising treatment for cervical cancer. However, no treatment for cervical cancer has yet been developed for clinical use or even for clinical trial. Besides the challenge in the development of a safe in vivo delivery system, the effectiveness of RNAi itself as a mono-therapy to treat cervical cancer patients, especially at late clinical and metastasis stages, is unknown. In addition, the existence of cervical CSCs and their sensitivity to RNAi-based treatments have not been investigated. These are challenging questions and will potentially provide strategic solutions for cancer therapy.

In this study, we firstly isolated sphere-forming cells (SFCs) from cervical cancer cell lines using a method that has been used to successfully isolate or enrich stem cells or CSCs from mammary tissues,27 primary tumors or cancer cell lines28, 29, 30, 31 Characterization of these SFCs showed CSC-like features, including higher tumorigenicity than parental HeLa cells, expression of biomarkers CD44high/CD24low and multi-drug resistance. We further demonstrated that these cells were sensitive to E6 oncogene silencing, as evidenced by the profound inhibition of cell growth and sphere formation. We also showed that this effect was likely through a reduction in the self-renewal ability modulated by transformation growth factor-beta (TGF-β).

Materials and methods

Cervical cancer cell lines and sphere culture

Cervical cancer cell lines HeLa (ATCC, CCL-2), CaSki (ATCC, CRL-1550), SiHa (ATCC, HTB-35) and C33A (ATCC, HTB-31) were purchased from American Type Culture Collection and maintained in complete Dulbecco's modified Eagle's medium (Invitrogen, Gladesville, NSW, Australia)+10% fetal calf serum, as described.15 The sphere culture medium was prepared as previously described.27 For the first passage of sphere culture, 2 × 104 cancer cells were seeded into a T75 flask with 20 ml of sphere culture medium. Cells were cultured in suspension for 4 days at 37 °C. An additional 10 ml of sphere culture medium was added to the culture on day 4 and the culture continued for another 4–5 days. Spheres were harvested by centrifugation at 300 g for 5 min, and sphere numbers counted after re-suspension in 5 ml medium. For the second or following passage sphere culture, the spheres were treated with 1:1 diluted 2.5% Trypsin-EDTA (Invitrogen) for 5 min at 37 °C and washed with sphere culture medium. Spherical cells were separated by repeated pipetting and counted. The separated SFCs were then passed through a cell strainer (40 μM, BD, Brisbane, QLD, Australia) and used for continuous sphere culture or other assays in low-adherence six-well plates (Sigma-Aldrich, Sydney, NSW, Australia).

Drug treatment, cell staining and fluorescence microscopy

Chemotherapeutic drugs cisplatin, doxyrubicin and epitoside were purchased from Sigma-Aldrich. Their working concentrations in HeLa cells were cisplatin (1–2 μg ml−1), doxyrubicin (0.0625 μg ml−1) and epitoside (0.5 μg ml−1), at which they kill >90% HeLa cells in 24–72 h. For fluorescence-activated cell sorting analysis, SFCs were harvested and dispersed from spheres. The cells were stained with antibodies to CD44 conjugated with fluorescein isothiocyanate and to CD24 conjugated with R-phycoerythrin (Invitrogen) at concentrations of 1:100 (V/V, 1 × 105 cells). The cells were washed two times with 1% fetal calf serum/phosphate-buffered saline, then fixed with 2% paraformaldehyde/phosphate-buffered saline for fluorescence-activated cell sorting analysis using Calibur or FACS Canto (BD). For fluorescence microscopy, a small portion of fixed cells was cytospun onto microscope slides using a Cytospin 4 (Thermo Shandon, Cheshire, UK) and mounted in Fluoroshield mount medium containing 4′,6-diamidino-2-phenylindole (Sigma-Aldrich) to stain cell nuclei.

Animal experiment

HeLa and HeLa-SFCs were trypsinized, washed and re-suspended in phosphate-buffered saline at different cell concentrations (2 × 105 to 2 × 107 cells ml−1). Female non-obese diabetic, severe combined immunodeficient (NOD/SICD) 6–8-week-old mice were used and subcutaneously injected (three mice per group) with 50 μl of cell suspension to the neck scruff. The tumors were monitored weekly (and size measured with a calipers) and dissected at 22 days post injection for final examination. The animal experiment was approved by the Animal Ethics Committee of Queensland University.

Colony-forming assay

HeLa cells and HeLa-SFCs were harvested as above and counted. A total of 100 cells were seeded in each well of 12-well plates with 1 ml complete Dulbecco's modified Eagle's medium; each group had three replicates. The cells were cultured for 7 days and then fixed with 1 ml 95% ethanol for 30 min at room temperature and stained with 0.5 ml 0.1% crystal violet for 5 min. The number of colonies in each well was counted and the data were expressed as mean±s.d.

Transduction of HeLa cells with lentiviral-shRNA

The production of lentiviral vectors carrying shRNA (LV-shRNA) and the transduction of HeLa cells with LV-shRNA were done as previously described.15 LV-shRNA 18E6-1 was shown to effectively silence HPV E6 and E7 genes,15 whereas the shRNA control LV-shRNA 16E7-2 was ineffective.26

Real-time RT-PCR

Total RNA extraction from transduced HeLa and HeLa-SFCs was done as instructed by the manufacturer using TRIzol reagent (Invitrogen). Reverse transcription reactions were performed with oligo-dT primer using the High Capacity cDNA RT Kit (Applied Biosystems, Melbourne, VIC, Australia). Real-time PCR was carried out with SYBR green master mixture (Promega, Sydney, NSW, Australia) using a Rotor-Gene RG-3000 (Corbett Research, Doncaster, VIC, Australia) with the program pre-heating at 95 °C for 10 min, followed by 40 cycles of 94 °C 15 s, 58 °C 30 s and 72 °C 45 s.

The primers were HPV 18E7: F 5′-AAAATGAAATTCCGGTTGA-3′, R 5′-GGCTGGTAAATGTTGATGAT-3′; HPV 18E6: F 5′-CTGTGCACGGAACTGAACAC-3′, R 5′-TGCAGCATGGGGTATACTGT-3′; TGF-β-1: F 5′-CAACAATTCCTGGCGATACC-3′, R 5′-GAACCCGTTGATGTCCACTT-3′; TGF-β-2: F 5′-GAGTGCCTGAACAACGGATT-3′, R 5′-TGCAGCAGGGACAGTGTAAG-3′; TGF-β-3: F 5′-GCAACTTGGAGGAGAACTGC-3′, R 5′-CTGTGGGTTGTGTCTGCACT-3′; leukemia-inhibitory factor (LIF): F 5′-CCCTGTCGCTCTCTAAGCAC-3′, R 5′-ATCCTGGACAAGGGTGAGTG-3′; H-Ras: F 5′-GTGGTCATTGATGGGGAGAC-3′, R 5′-ACGTCATCCGAGTCCTTCAC-3′; and K-Ras: F 5′-TGT CAAGCTCAGCACAATCTG-3′, R 5′-GGTAGGGAGGCAAGATGACA-3′. The internal control was 18S rDNA (F: 5′-CCATCGAACGTCTGCCCTA-3′, R: 5′-TCACCCGTGGTCACCATG-3′) and it was also used to normalize gene expression.

Data analysis

Data collected from each (experimental and control) group were expressed as mean±s.d. One-way analysis of variance and unpaired Student's t-test (GraphPad Prism 5 program) were used to analyze the differences between groups and distinguish the significant differences (two-tailed, P<0.05) between experimental and control groups.

Results

Sphere culture of cervical cancer cell lines

We tested four cervical cancer cell lines, HeLa, SiHa, CasKi and C33A, for their sphere-formation abilities; only HeLa and SiHa were able to form cancer cell spheres. We therefore selected HeLa cells for the following studies. When cultured in sphere culture conditions, some HeLa cells formed spheres at day 4 while most HeLa cells adhered to the wall of the culture flask (Figure 1a). At day 9, the spheres became very compact cell clusters (Figures 1b and c). Previous studies showed that breast CSCs from mammo-sphere cells exhibited surface markers of high CD44 and low CD24.28, 32 As cervical cancer is also a female epithelial cancer, we examined CD44 and CD24 expression in HeLa-SFCs. The results showed that HeLa-SFCs were CD44 positive and CD24 low, whereas parental HeLa cells were CD44 negative and CD24 high (Figure 1d), the same pattern as in breast CSCs.

Figure 1
figure1

Morphology of HeLa sphere and HeLa-SFCs. (a) HeLa spheres were cultured in sphere culture medium for 4 days, whereas some other HeLa cells still survived as adhesive cells. (b) HeLa spheres at day 8 of culture; they were bigger spheres with a compacted cell cluster inside. (c) HeLa spheres at day 10 of culture; the inside of the sphere contains hundreds of SFCs. (d) HeLa-SFCs were highly positive for CD44 and expressed a low level of CD24, whereas their parental HeLa cells were CD44 negative and expressed a high level of CD24 as analyzed by FACS. The fluorescent images show the CD44 staining on the cell surface with an anti-CD44 antibody conjugated with fluorescein isothiocyanate.

Tumorigenicity and drug resistance of HeLa SFCs

To examine whether HeLa-SFCs are more tumorigenic than parental HeLa cells, we subcutaneously injected the cells to NOD/SCID mice at different dosages (Table 1). HeLa cells were also injected as controls. In the HeLa-SFC group, 1 × 104 cells could form tumors in 3 weeks, whereas 1 × 106 parental HeLa cells were needed to grow similar-sized tumors in the same time frame (Table 1), suggesting HeLa-SFCs are more tumorigenic (100 times) than their parental HeLa cells and that sphere culture is a means of enriching tumorigenic cancer cells. To test whether HeLa-SFCs are also multi-drug resistant, we cultured the cells in the presence of three clinically used chemotherapy drugs at concentrations that are cytotoxic for HeLa cells. The presence of the three drugs did not affect SFC sphere formation or growth either at passage 1 or at passage 5 (Figure 2a), suggesting HeLa-SFCs are resistant to these drugs. When HeLa-SFCs were cultured in complete Dulbecco's modified Eagle's medium containing fetal calf serum, these cells quickly adopted the morphology of HeLa cells (Figure 2b) and became sensitive to the same drug treatments (Figure 2c). This result suggests that after differentiation, HeLa-SFCs regain their HeLa cell properties. In addition, when they were cultured in complete Dulbecco's modified Eagle's medium, HeLa-SFCs formed more and bigger colonies than HeLa cells (Figure 2d and e), indicating SFCs have a greater ability to grow, and that they grow faster than HeLa cells. Taken together, the above data suggest that HeLa-SFCs have many of the characteristics of CSCs.

Table 1 Xeno-transplant tumors in NOD/SCID mice at 22 days after injection of HeLa-SFCs
Figure 2
figure2

Characterization of HeLa-SFCs. (a) Multi-drug resistances of HeLa-SFCs; they were cultured in sphere culture medium containing working concentrations of cisplatin (CIS), doxyrubicin (DOX) and epitoside (EPI), respectively. Their sphere formations were measured at sphere culture passage 1 (P1) and 5 (P5). The HeLa-SFCs cultured without the above drugs were used as negative control. (b) When HeLa-SFCs were cultured in complete Dulbecco's modified Eagle's medium (DMEM) containing fetal calf serum, they quickly grew as adhesive cells and grew into the morphology of the parental HeLa cells. (c) When HeLa-SFCs grew as adhesive cells they became sensitive to CIS treatment and most cells were killed by the working concentration (1 × ) of CIS. The crystal violet (blue) stained only the living and adhesive cells in the wells. The dead cells were floating and washed away before the staining. (d, e) HeLa-SFCs were plated for colony-forming assay in complete DMEM; they grew bigger colonies (d) and formed more colonies (e) than the parental HeLa cells. ***P<0.001.

Silencing E6/E7 affects cell growth and sphere formation of HeLa-SFCs

As HeLa-SFCs showed the characteristics of CSCs, we then investigated how E6/E7 gene silencing would affect their growth and sphere formation. We first examined the expression of E6 and E7 by HeLa-SFCs. As expected, HeLa-SFCs expressed both E6 and E7. However, to our surprise, HeLa-SFCs expressed 6.9-fold more E6 than the parent HeLa cells, whereas there was no significant difference in E7 expression (Figure 3a). This result indicates that tumorigenic cancer cells may have an oncogene expression profile different from that of differentiated cancer cells and that the investigation of the effect of E6 silencing in HeLa-SFCs is more interesting.

Figure 3
figure3

The effect of E6/E7 silencing on HeLa-SFC growth and sphere formation. (a) HPV E6 and E7 expression levels in HeLa and HeLa-SFCs were measured using real-time PCR and the data were representative for at least two independent assays. (b) Sphere growth of HeLa cells transduced with LV-18E6-1 and -16E7-2 in sphere culture; the sphere growth of LV-18E6-1 (18E6-1-sphere) was obviously inhibited compared with control LV-16E7-2 (16E7-2-sphere), as evidenced by their smaller-sized spheres. (c) The transduced HeLa-SFCs were analyzed by FACS for their enhanced green fluorescent protein (eGFP) expression (lentiviral vector carrying an eGFP gene); the majority of them were positive. (d) The transduced HeLa-SFCs were measured for E6 and E7 gene expression using real-time PCR. Compared with control 16E7-2, 18E6-1-transduced SFCs had significantly lower levels of E6 and E7 expression. *P<0.05; **P<0.01.

We treated HeLa-SFCs with lentiviral-shRNA (18E6-1) targeting HPV E6 (as well as E7, as the two genes are expressed bicistronically15). The data showed that this significantly inhibited sphere formation (sphere numbers) and also cell growth (cell number within each sphere; Figure 3b and Table 2). As sphere formation and growth represent the self-renewal ability of the cells, these data suggest that E6/E7 may have a profound impact on SFC self-renewal. Thus, further investigations into the molecular mechanism are required. We then measured enhanced green fluorescent protein expression in the SFCs by flow cytometry. Figure 3c shows that the majority of SFCs were green fluorescent protein positive, confirming that the SFCs were transduced by the lentiviral vector. To confirm the gene-silencing effect, we also measured E6/E7 expression levels in transduced HeLa-SFCs. The results proved that transduced HeLa-SFCs had lower E6 and E7 levels, compared with the control (Figure 3d). It was observed that the lentiviral vector itself had some non-specific effects on sphere growth and formation (Table 2, 16E7-2). However, compared with the non-specific LV-shRNA control (16E7-2), 18E6-1 treatment had significant effects on both sphere formation (total spheres in Table 2) and growth (total SFCs in Table 2). The data also showed that the inhibition of cell growth (P<0.001) was even more profound than sphere formation (P<0.05, Table 2), suggesting that the gene silencing of E6/E7 in HeLa SFCs has a major inhibitory effect on SFC growth (smaller spheres), with a less-degree effect on causing cell death (fewer spheres).

Table 2 Silencing of E6/E7 affects sphere formation and cell growth (ratio)

Silencing E6/E7 downregulates self-renewal gene TGF-β expression in HeLa-SFCs

TGF-β and LIF are two well-documented self-renewal genes in stem cells, with TGF-β also shown to increase the self-renewal ability of CSCs.33, 34 To understand whether silencing of E6/E7 would affect HeLa-SFC growth through self-renewal pathways, we examined the expression of TGF-β and LIF genes in transduced HeLa-SFCs. Real-time PCR results showed that expression of all three TGF-β isoforms was significantly decreased (Figure 4a) but the LIF gene was not affected (Figure 4b), suggesting that the self-renewing ability of HeLa-SFCs is associated with the TGF-β pathway. The TGF-β family has a very close relationship with human Ras genes and they interact with each other (reviewed by Grusch et al35). To further confirm the downregulation of TGF-β, we measured H-Ras and K-Ras expression in LV-shRNA-transduced HeLa-SFCs. The results showed that both H-Ras and K-Ras were significantly decreased in the 18E6-1 group compared with the control 16E7-2 (Figure 4c), supporting the reduction of TGF-β after E6/E7 silencing with 18E6-1 shRNA.

Figure 4
figure4

Self-renewal and Ras gene expressions in transduced HeLa-SFCs. The expression levels of three isoforms of TGF-β (TGF-β-1, -2 and -3) were measured in transduced HeLa-SFCs using real-time PCR; all of them decreased significantly in LV-18E6-1-treated cells compared with the control LV-16E7-2 (a). Another self-renewal gene, LIF, was similarly measured in the same cells but no change was observed (b). The expression levels of human Ras genes (including H-Ras and K-Ras) were also measured and they all significantly decreased in 18E6-1-treated cells, compared with the control (c). **P<0.01; ***P<0.001.

Discussion

Because of the special role of CSCs in tumor initiation, maintenance and drug resistance, they are considered as important targets for future cancer treatments. However, as these cells can easily gain resistance to chemo/radiotherapies, these conventional methods will not be effective for further treatment of these cells. RNAi-based therapies have shown great potential for the treatment of diseases such as viral infections, genetic disorders and cancers. We thus investigated whether silencing oncogenes expressed by CSCs would significantly affect their growth and self-renewal ability. To achieve this, we first isolated and characterized cervical cancer HeLa-SFCs and found that they exhibited CSC features. We then showed that they expressed 6.9 times more HPV E6 than their parental HeLa cells, and that silencing the E6 oncogene had a profound effect on their cell growth and self-renewal ability. We also showed that the inhibitory effect was associated with the downregulation of self-renewal gene TGF-β and other oncogenes such as Ras. To our knowledge, this is the first report on CSC oncogene silencing using RNAi. The data suggest the potential of RNAi-based therapy to target CSCs.

Previously, we showed that silencing of HPV E6/E7 with LV-18E6-1 could effectively inhibit HeLa cell growth in vitro and in vivo and could significantly decrease tumor nodules in a mouse lung metastasis model.15 According to the current study, the action of LV-18E6-1 in the previous study may have been due to its ability to target cervical CSCs; this may be especially important in the lung metastasis model, as CSCs are reported to be responsible for initiating metastatic tumors.30 These data suggest that RNAi-based therapy may be used as a sole therapy, as it targets both CSCs and differentiated cancer cells by silencing their oncogenes. Alternatively, it could be used in combinational therapies with chemotherapy or radiotherapy to further target the resistant CSCs and improve treatment outcomes.

To confirm the gene-silencing effect in SFCs, we measured HPV E6 and E7 expression levels in transduced HeLa-SFCs (Figure 3d). Compared with the inhibitory effect on sphere formation and cell growth, the decrease of E6 and E7 levels was not as profound as expected, though the statistical analysis showed significance (P<0.05 or 0.01; Figure 3d). One possibility is the sample was taken after the period of sphere culture (which normally takes 8–10 days) following transduction, and this time point might have passed the peak gene-silencing time; according to our previous study, the peak protein-inhibitory effect of 18E6-1 shRNA was around 1 week.15 Another possibility is that the shRNA might silence the gene expression by partially blocking protein translation rather than cleaving the mRNA.

The TGF-β family regulates cell survival, proliferation, differentiation and adhesion.36, 37 TGF-β is also reported to be an important self-renewal factor for embryonic stem cells,38 to promote CSC self-renewal33 and to increase the self-renewal ability of glioma-initiating cells.34 In the current study, we observed that after silencing of E6/E7, expression of all three isoforms of TGF-β was significantly decreased in HeLa-SFCs, whereas that of another self-renewal gene, LIF, remained unchanged. This result suggests that E6/E7 silencing has a specific effect on expression of the self-renewal gene, TGF-β. As TGF-β isoforms are reported to have slightly different roles in different cancers,39 the result reveals that the silencing of E6/E7 may activate a general mechanism that downregulates the expression of all three isoforms of TGF-β. Although the mechanisms are not clear, E6 and E7 are known to promote cancer cell growth through the PI3K/Akt pathway,40, 41 and reduce apoptosis through the p53 pathway.20, 42 TGF-β also promotes CSC growth and one of its downstream targets is PI3K/Akt.43 We postulate that downregulation of the PI3K/Akt pathway after E6/E7 silencing is the link that indirectly affects TGF-β expression, but this needs further investigation.

To prove the downregulation of TGF-β after E6/E7 silencing, we measured expression of both H-Ras and K-Ras genes, as Ras is a downstream component of the TGF-β pathway.35 The result confirmed that expression of both genes was decreased in transduced HeLa-SFCs. Furthermore, during epithelial–mesenchymal transition, the TGF-β family and Ras have an even closer relationship as TGF-β has been shown to promote tumorigenicity only in Ras-expressing cells.44 These data further support the conclusion that TGF-β regulates Ras gene expression, especially in the tumorigenic epithelial–mesenchymal transition cancer cells, and the decrease of Ras gene expression is due to the downregulation of TGF-β. The decreased Ras expression could then interfere with the growth of HeLa-SFCs after E6/E7 silencing. Although a previous study showed that TGF-β regulates self-renewal of glioma-initiating cells (or CSCs) through LIF,34 we did not observe similar changes in LIF, suggesting that different cancer models may utilize different self-renewal pathways.

Since the first report of sphere culture in 2003,27 this method has been used by many laboratories for isolating CSCs from different cancers, with these cells shown to be multi-drug resistant, more tumorigenic and more metastatic.28, 45, 46 An advantage of this method over other methods is that it does not involve the cell-sorting procedure and limits the damages to the cells. Furthermore, as this method is selective for self-renewal, a definitive character of stem cells,47 there is a better chance to isolate CSCs. As shown in the present study, this method can also be used to isolate CSCs from cancer cell lines, which has some advantages over isolation from primary tumors. For example, cancer cell lines can be easily treated with chemotherapeutic drugs in vitro to increase sphere formation rates.46 Our data suggest that cisplatin-treated HeLa cells can increase the sphere formation to 27%, whereas doxyrubicin-treated HeLa can reach 96% (unpublished data). Another important advantage is that isolating CSCs from cancer cell lines may avoid potential contamination with normal stem cells.

To summarize, this study shows a new approach to target CSCs for future cancer therapy. At least in this cervical cancer model, targeting oncogenes E6/E7 has profound effects on cancer stem-like cell growth and self-renewal ability, which are mediated through the TGF-β pathway. The sphere culture method can be effective in enriching CSCs from cancer cell lines, and this can be further enhanced by treating the cell lines with chemotherapeutic drugs. Further investigations in this direction may lead to a better understanding of CSCs and, ultimately, better cancer treatments by targeting CSCs.

Abbreviations

CSCs:

cancer stem cells

HPV:

human papillomavirus

shRNA:

short-hairpin RNA

SFCs:

sphere-forming cells

TGF-β:

transformation growth factor-beta

LIF:

leukemia-inhibitory factor

References

  1. 1

    Visvader JE, Lindeman GJ . Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer 2008; 8: 755–768.

  2. 2

    Nakai E, Park K, Yawata T, Chihara T, Kumazawa A, Nakabayashi H et al. Enhanced MDR1 expression and chemoresistance of cancer stem cells derived from glioblastoma. Cancer Invest 2009; 27: 901–908.

  3. 3

    Lou H, Dean M . Targeted therapy for cancer stem cells: the patched pathway and ABC transporters. Oncogene 2007; 26: 1357–1360.

  4. 4

    Kondo T . Stem cell-like cancer cells in cancer cell lines. Cancer Biomark 2007; 3: 245–250.

  5. 5

    Harper SQ, Staber PD, He X, Eliason SL, Martins IH, Mao Q et al. RNA interference improves motor and neuropathological abnormalities in a Huntington's disease mouse model. Proc Natl Acad Sci USA 2005; 102: 5820–5825.

  6. 6

    Xia H, Mao Q, Eliason SL, Harper SQ, Martins IH, Orr HT et al. RNAi suppresses polyglutamine-induced neurodegeneration in a model of spinocerebellar ataxia. Nat Med 2004; 10: 816–820.

  7. 7

    Qin XF, An DS, Chen IS, Baltimore D . Inhibiting HIV-1 infection in human T cells by lentiviral-mediated delivery of small interfering RNA against CCR5. Proc Natl Acad Sci USA 2003; 100: 183–188.

  8. 8

    Jacque JM, Triques K, Stevenson M . Modulation of HIV-1 replication by RNA interference. Nature 2002; 418: 435–438.

  9. 9

    Schomber T, Kalberer CP, Wodnar-Filipowicz A, Skoda RC . Gene silencing by lentivirus-mediated delivery of siRNA in human CD34+ cells. Blood 2004; 103: 4511–4513.

  10. 10

    Vaishnaw AK, Gollob J, Gamba-Vitalo C, Hutabarat R, Sah D, Meyers R et al. A status report on RNAi therapeutics. Silence 2010; 1: 14.

  11. 11

    Borkhardt A . Blocking oncogenes in malignant cells by RNA interference--new hope for a highly specific cancer treatment? Cancer Cell 2002; 2: 167–168.

  12. 12

    Sledz CA, Williams BR . RNA interference in biology and disease. Blood 2005; 106: 787–794.

  13. 13

    Jiang M, Milner J . Selective silencing of viral gene expression in HPV-positive human cervical carcinoma cells treated with siRNA, a primer of RNA interference. Oncogene 2002; 21: 6041–6048.

  14. 14

    Putral LN, Bywater MJ, Gu W, Saunders NA, Gabrielli BG, Leggatt GR et al. RNA interference against human papillomavirus oncogenes in cervical cancer cells results in increased sensitivity to cisplatin. Mol Pharmacol 2005; 68: 1311–1319.

  15. 15

    Gu W, Putral L, Hengst K, Minto K, Saunders NA, Leggatt G et al. Inhibition of cervical cancer cell growth in vitro and in vivo with lentiviral-vector delivered short hairpin RNA targeting human papillomavirus E6 and E7 oncogenes. Cancer Gene Ther 2006; 13: 1023–1032.

  16. 16

    Yamato K, Fen J, Kobuchi H, Nasu Y, Yamada T, Nishihara T et al. Induction of cell death in human papillomavirus 18-positive cervical cancer cells by E6 siRNA. Cancer Gene Ther 2006; 13: 234–241.

  17. 17

    Faltus T, Yuan J, Zimmer B, Kramer A, Loibl S, Kaufmann M et al. Silencing of the HER2/neu gene by siRNA inhibits proliferation and induces apoptosis in HER2/neu-overexpressing breast cancer cells. Neoplasia 2004; 6: 786–795.

  18. 18

    Wang YH, Liu S, Zhang G, Zhou CQ, Zhu HX, Zhou XB et al. Knockdown of c-Myc expression by RNAi inhibits MCF-7 breast tumor cells growth in vitro and in vivo. Breast Cancer Research 2005; 7: R220–R228.

  19. 19

    Liu TG, Yin JQ, Shang BY, Min Z, He HW, Jiang JM et al. Silencing of hdm2 oncogene by siRNA inhibits p53-dependent human breast cancer. Cancer Gene Ther 2004; 11: 748–756.

  20. 20

    Narisawa-Saito M, Kiyono T . Basic mechanisms of high-risk human papillomavirus-induced carcinogenesis: roles of E6 and E7 proteins. Cancer Sci 2007; 98: 1505–1511.

  21. 21

    zur Hausen H . Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2002; 2: 342–350.

  22. 22

    Chang JT, Kuo TF, Chen YJ, Chiu CC, Lu YC, Li HF et al. Highly potent and specific siRNAs against E6 or E7 genes of HPV16- or HPV18-infected cervical cancers. Cancer Gene Ther 2010; 17: 827–836.

  23. 23

    Jonson AL, Rogers LM, Ramakrishnan S, Downs Jr LS . Gene silencing with siRNA targeting E6/E7 as a therapeutic intervention in a mouse model of cervical cancer. Gynecol Oncol 2008; 111: 356–364.

  24. 24

    Sima N, Wang W, Kong D, Deng D, Xu Q, Zhou J et al. RNA interference against HPV16 E7 oncogene leads to viral E6 and E7 suppression in cervical cancer cells and apoptosis via upregulation of Rb and p53. Apoptosis 2008; 13: 273–281.

  25. 25

    Tang S, Tao M, McCoy Jr JP, Zheng ZM . Short-term induction and long-term suppression of HPV16 oncogene silencing by RNA interference in cervical cancer cells. Oncogene 2006; 25: 2094–2104.

  26. 26

    Gu W, Cochrane M, Leggatt GR, Payne E, Choyce A, Zhou F et al. Both treated and untreated tumors are eliminated by short hairpin RNA-based induction of target-specific immune responses. Proc Natl Acad Sci USA 2009; 106: 8314–8319.

  27. 27

    Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev 2003; 17: 1253–1270.

  28. 28

    Yu F, Yao H, Zhu P, Zhang X, Pan Q, Gong C et al. let-7 regulates self renewal and tumorigenicity of breast cancer cells. Cell 2007; 131: 1109–1123.

  29. 29

    Yeung TM, Gandhi SC, Wilding JL, Muschel R, Bodmer WF . Cancer stem cells from colorectal cancer-derived cell lines. Proc Natl Acad Sci USA 2010; 107: 3722–3727.

  30. 30

    Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, Finetti P et al. Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a distinct molecular signature. Cancer Res 2009; 69: 1302–1313.

  31. 31

    Li F . Every single cell clones from cancer cell lines growing tumors in vivo may not invalidate the cancer stem cell concept. Mol Cells 2009; 27: 491–492.

  32. 32

    Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF . Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 2003; 100: 3983–3988.

  33. 33

    Singh A, Settleman J . EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer. Oncogene 2010; 29: 4741–4751.

  34. 34

    Penuelas S, Anido J, Prieto-Sanchez RM, Folch G, Barba I, Cuartas I et al. TGF-beta increases glioma-initiating cell self-renewal through the induction of LIF in human glioblastoma. Cancer Cell 2009; 15: 315–327.

  35. 35

    Grusch M, Petz M, Metzner T, Ozturk D, Schneller D, Mikulits W . The crosstalk of RAS with the TGF-beta family during carcinoma progression and its implications for targeted cancer therapy. Curr Cancer Drug Targets 2010; 10: 849–857.

  36. 36

    Kang JS, Liu C, Derynck R . New regulatory mechanisms of TGF-beta receptor function. Trends Cell Biol 2009; 19: 385–394.

  37. 37

    Heldin CH, Landstrom M, Moustakas A . Mechanism of TGF-beta signaling to growth arrest, apoptosis, and epithelial-mesenchymal transition. Curr Opin Cell Biol 2009; 21: 166–176.

  38. 38

    Watabe T, Miyazono K . Roles of TGF-beta family signaling in stem cell renewal and differentiation. Cell Res 2009; 19: 103–115.

  39. 39

    Bellone G, Carbone A, Tibaudi D, Mauri F, Ferrero I, Smirne C et al. Differential expression of transforming growth factors-beta1, -beta2 and -beta3 in human colon carcinoma. Eur J Cancer 2001; 37: 224–233.

  40. 40

    Charette ST, McCance DJ . The E7 protein from human papillomavirus type 16 enhances keratinocyte migration in an Akt-dependent manner. Oncogene 2007; 26: 7386–7390.

  41. 41

    Menges CW, Baglia LA, Lapoint R, McCance DJ . Human papillomavirus type 16 E7 up-regulates AKT activity through the retinoblastoma protein. Cancer Res 2006; 66: 5555–5559.

  42. 42

    Scheffner M, Whitaker NJ . Human papillomavirus-induced carcinogenesis and the ubiquitin-proteasome system. Semin Cancer Biol 2003; 13: 59–67.

  43. 43

    Assinder SJ, Dong Q, Kovacevic Z, Richardson DR . The TGF-beta, PI3K/Akt and PTEN pathways: established and proposed biochemical integration in prostate cancer. Biochem J 2009; 417: 411–421.

  44. 44

    Safina AF, Varga AE, Bianchi A, Zheng Q, Kunnev D, Liang P et al. Ras alters epithelial-mesenchymal transition in response to TGFbeta by reducing actin fibers and cell-matrix adhesion. Cell Cycle 2009; 8: 284–298.

  45. 45

    Levina V, Marrangoni AM, DeMarco R, Gorelik E, Lokshin AE . Drug-selected human lung cancer stem cells: cytokine network, tumorigenic and metastatic properties. PLoS One 2008; 3: e3077.

  46. 46

    Bertolini G, Roz L, Perego P, Tortoreto M, Fontanella E, Gatti L et al. Highly tumorigenic lung cancer CD133+ cells display stem-like features and are spared by cisplatin treatment. Proc Natl Acad Sci USA 2009; 106: 16281–16286.

  47. 47

    Dean M, Fojo T, Bates S . Tumour stem cells and drug resistance. Nat Rev Cancer 2005; 5: 275–284.

Download references

Acknowledgements

We would like to acknowledge the financial supports of NHMRC of Australia (Peter Doherty Fellowship and Travelling Award to WG, project grant ID631402 to NM) and Australian Cancer Foundation. We would also like to thank Prof Judy Lieberman and Dr Fabio Petrocca at Harvard Medical School for their support and technique help. We also thank Dr Barbara Rolfe at the University of Queensland for reading and providing useful suggestions to the manuscript.

Author information

Correspondence to W Gu or C Yu.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Gu, W., Yeo, E., McMillan, N. et al. Silencing oncogene expression in cervical cancer stem-like cells inhibits their cell growth and self-renewal ability. Cancer Gene Ther 18, 897–905 (2011) doi:10.1038/cgt.2011.58

Download citation

Keywords

  • cancer stem cells
  • sphere-forming cells
  • RNA interference
  • HeLa cells
  • cervical cancer
  • HPV E6/E7
  • self-renewal gene
  • TGF-β
  • lentiviral vector

Further reading