Down-regulation of Sp1 suppresses cell proliferation, clonogenicity and the expressions of stem cell markers in nasopharyngeal carcinoma

The reagents used were as follows: rabbit polyclonal antibodies against Sp1 (#07-645, Millipore, Temecula, CA, USA), p27 (#60-645, Millipore), CDK4 (ab137818, Abcam, Cambridge, MA, USA); mouse monoclonal antibodies (mAb) against p21 (sc-6246, Santa Cruz, CA, USA), α-tublin (T6199, Sigma-Aldrich, St Louis, MO, USA), GAPDH (sc-32233, Santa Cruz). All other reagents were obtained from Sigma-Aldrich, unless otherwise indicated.

The paraffin-embedded NPC specimens (n = 76) were obtained from Cancer Center of Sun Yat-sen University in Guangzhou, People’s Republic of China. Written informed consent was obtained from each patient, and the study was approved by the Institute Research Ethics Committee at the Cancer Center. The clinical characteristics of the NPC patients were described in Table 1. All cases were histologically confirmed.

Normal primary NPECs (NPECw and NPEC01) were established as described previously [34], and cultured in keratinocyte/serum-free medium (Invitrogen, Carlsbad, CA, USA). The NPC cell lines (CNE2, HNE1, HONE1, C666-1 and HK1) were cultured in RPMI 1640 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT), and in a humidified 5% CO2 incubator at 37°C.

The siRNA oligoribonucleotides were purchased from Dharmacon (Rockford, USA) and Ribobio (Ribobio, Guangzhou, China). The siRNAs targeting the mRNA of human Sp1 [GenBank: NM_138473.2] was denoted as siSp1-1# (l-026959-00-0003) and siSp1-2# (siG0898155658). The negative control (NC) RNA duplex for the siRNA was nonhomologous to any human genome sequences, and was indicated as siNC (D-001220-01-20). CNE2 and HNE1 cells (1 × 10⁵) were seeded on 6-well plates for 16 h, followed by transfection with 50 nM of the RNA duplex and 5 μL of Lipofectamine RNAiMAX (Invitrogen, Grand Island, NY, USA), according to the manufacturer’s instructions. After 72 h, cells were washed with PBS and harvested for further experiments.

Total RNA from NPC cell lines or tissue specimens was extracted using the TRIzol reagent (Invitrogen, Grand Island, NY, USA) according to the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized from 1 μg of the total RNA using a reverse transcriptase kit (Invitrogen). The mRNA level was determined by quantitative Real-time RT-PCR (qRT-PCR) using the Power SYBR Green qPCR SuperMix-UDG (Invitrogen, Grand Island, NY, USA) and was analyzed on an ABIPrism-7500 Sequence Detector System (ABI, applied biosystems, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or β-actin was used as an internal control. The primers are listed in Table 2. Relative expression of the indicated gene was normalized to GAPDH or β-actin expression, which yielded a 2^-△△Ct value. All reactions were run in triplicate.

Western blot was performed as previously described [32]. Briefly, cells were harvested and lysed in RIPA buffer containing protein inhibitor cocktail (F. Hoffmann-La Roche Ltd, Basel, Switzerland). The protein concentration was determined by the BCA method (Pierce, Rockford, IL). The protein (10 μg) was electrophoretically separated in 9 or 12% SDS polyacrylamide gels, and transferred to polyvinylidene difluoride membranes (Millipore). After probing with the indicated antibodies, the signals were detected using enhanced chemiluminescence (ECL) (Amersham Pharmacia Biotech, Piscataway, NJ). The same membranes were stripped and reprobed with mouse monoclonal antibodies against GAPDH or α-tubulin to confirm equal loading of the samples.

Immunohistochemical staining (IHC) was performed as previously described [34]. Briefly, paraffin-embedded samples were cut into 4 μm sections and placed on polylysine-coated slides. Paraffin sections were baked for 2 h at 58°C, de-paraffinized in xylene, rehydrated through graded ethanol, quenched for endogenous peroxidase activity in 0.3% hydrogen peroxide for 10 min, and processed for antigen retrieval by high pressure cooking in citrate antigen retrieval solution (pH = 6.0) for about 5 min. Sections were incubated at 4°C overnight with rabbit polyclonal antibodies against Sp1 (Millipore, 1:3200) in a moist chamber. Immunostaining was performed using the Zymed Histostain TM­Plus Kits (Zymed, South San Francisco, CA, USA), which resulted in a brown-colored precipitate at the antigen site. Subsequently, sections were counterstained with hematoxylin (Zymed Laboratories, South San Francisco, CA) and mounted in a non-aqueous mounting medium. All runs included a no primary antibody control. The immunohistochemically stained tissue sections were scored separately by two pathologists blinded to the clinical parameters. The staining intensity was scored as 0 (negative), 1 (weak), 2 (medium) or 3 (strong). Extent of staining was scored as 0 (0-10%), 1 (10-25%), 2 (26-50%) and 3 (51%-100%) according to the percentages of the positive staining areas in relation to the whole carcinoma area. The product of the intensity and extent score was used as the final staining score (0–9) for Sp1. Tumors having a final staining score of < 4.5 were considered to be low and those with score of ≥ 4.5 were considered to be high.

Forty-eight hours after transfection with siSp1-1#, siSp1-2# or siNC, CNE2 and HNE1 cells were plated in triplicate at 2000 cells per well in 96-well plates and maintained in RPMI containing 10% FBS. At the indicated time points, 10 μL of 5 mg/mL MTT (3- (4, 5-dimethylthiazol-a-yl)-2, 5-diphenyl tetrazolium bromide) was added to each well, and cells were incubated for 4 h at 37°C. The media was then discarded, and 200 μl of DMSO was added to each well for 20 min at RT to dissolve the precipitates. Absorbance was measured using the Spectramax M5 (Molecular Devices, CA, USA) at a wavelength of 570 nm.

Forty-eight hours after transfection with siSp1-1#, siSp1-2# or siNC, CNE2 and HNE1 cells were plated in triplicate at 200 cells per well in six-well plates and maintained in RPMI containing 10% FBS for 10 days. Colonies were fixed with methanol and stained with 0.1% crystal violet in 20% methanol, and then were photographed and counted. All visible colonies were quantified.

Forty-eight hours after transfection with the indicated RNA duplex, CNE2 and HNE1 cells (1 × 10⁴) that were suspended in 0.33% top agar (in RPMI with 10% FBS) were plated onto 0.55% base agar (in RPMI with 5% FBS) in six-well plates for 14 days. The colonies with the diameter bigger than 10 μm were counted.

Forty-eight hours after transfection with the indicated RNA duplex, CNE2 cell (5 × 10²) that were suspended in tumor sphere medium containing serum-free DMEM/F-12, N₂ supplement, 10 ng/mL human recombinant bFGF and 10 ng/mL EGF were plated in six-well plates (ultra low attachment surface) for 7 days. Colonies were photographed at 200× magnification and were counted at 40× magnification.

To explore the significance of Sp1 in the development of nasopharyngeal carcinoma, the mRNA and protein levels of Sp1 were first examined in the normal primary nasopharyngeal epithelial cells (NPECs) and NPC cell lines. Both qRT-PCR and Western blot showed that the level of Sp1 in the four NPC cell lines (HNE1, HONE1, CNE2 and C666-1) was obviously higher than the normal primary NPECs (NPECw and NPEC01) (Figure 1). The expression and subcellular localization of Sp1 in 76 NPC tissues were then determined by immunohistochemistry. As shown in Figure 2, Sp1 was expressed predominantly in the nuclei of NPC cells of the tumor regions (T), but demonstrated weak signal in the NPEC of the non-tumor regions (N). In addition, Sp1 was present in some infiltrating lymphocytes of both the non-tumor and tumor regions of NPC patients. To evaluate the clinical significance of Sp1 in nasopharyngeal carcinoma, the correlation between Sp1 level and the clinical variations was analyzed using χ ² test. As shown in Table 3, the expression of Sp1 was significantly associated with the clinical stage (P = 0.00036), T (P = 0.002) and N (P = 0.034) classification. There was no significant correlation between the expression level of Sp1 and age, gender, M classification as well as histologic classification in NPC patients. These results suggested that the increase in intratumoral Sp1-positive NPC cells was associated with NPC progression.

To investigate the potential roles of Sp1 in tumorigenesis, the effect of Sp1 on the cell growth was first evaluated. CNE2 and HNE1 cells were cultured in the presence of 100 nM mithramycin A (MITA), a FDA-approved chemotherapeutic anticancer drug that inhibits transcriptional activity of Sp1 by competitively binding to the Sp1-binding sites. The cell viability was determined by MTT on the indicated times. Compared with DMSO-treated cells, the cell viability was significantly suppressed in MITA-treated CNE2 and HNE1 cells (Figure 3A).

Next, we assessed the effect of Sp1 knockdown on the cell growth of NPC cells. HNE1 and CNE2 cells were transfected with the two distinct siRNA duplexes against Sp1, and then subjected to Western blot. The level of Sp1 was nearly abolished in siSp1 transfectants (Figure 3B). Compared to the control, siSp1 transfectants displayed significantly lower proliferation, suggesting the growth-promoting role of Sp1 in NPC cells (Figure 3C).

To assess whether cell cycle progression was involved in Sp1-promoted cell growth, CNE2, HNE1 and HK1 cells were transfected with the indicated RNA duplex for 48 h, followed by flow cytometry analysis for the cell cycle distribution of initially asynchronous cells. Knockdown of Sp1 showed a marked induction of G1 arrest in CNE2, HNE1 and HK1 cells, whereas had no effect on apoptosis as evaluated by sub-G1 peak (Figure 4A and B). The G1/S-phase transition has been characterized as a balance of cyclins, CDKs and Cip1 proteins. Cyclins promote S phase, whereas p27 and p21 maintain cells in G1 arrest. CDK4, p27 and p21 have been reported to regulate the G1/S-phase transition and are potentially regulated by Sp1 [35]–[37]. To study whether Sp1 promotes cell cycle through the expression regulation of these molecules in nasopharyngeal carcinoma, CNE2, HNE1 and HK1 cells were transfected with siSp1 and siNC for 48 h, followed by analysis for the levels of Sp1, CDK4, p27 and p21. Either qRT-PCR or Western blot demonstrated that Sp1 knockdown obviously up-regulated the levels of p27 and p21, but had no effect on the expression levels of CDK4 (Figure 4C and D), indicating Sp1 knockdown blocked the G1/S transition through induction of p27 and p21 in NPC cells.

We then analyzed the effect of Sp1 on the clonogenicity and anchorage-independent growth of NPC cells. CNE2 and HNE1 cells were transfected with the indicated RNA duplex, and then allowed to grow at very low density or grow in soft agar. The siSp1-transfectants displayed obviously fewer and smaller colonies than the siNC-transfectants, as shown by a significant reduction in both the size and the number of colonies (Figure 5A and B). In addition, knockdown of Sp1 significantly suppressed the capacity of NPC cells to grow in soft agar (Figure 5C and D). These results indicated that Sp1 silencing exerted a suppressive role on the colony formation and anchorage-independent growth of NPC cells.

Stem cells can be expanded as sphere-like cellular aggregates in the sphere medium containing EGF and bFGF. To further determine the role of Sp1 silencing in the stem-cell like phenotype of NPC cells, sphere formation potential was monitored. CNE2 cells transfected with the indicated siRNA duplexes were cultivated in the sphere medium for 7 days. The sphere formation number was significantly lower in Sp1 silenced CNE2 cells (Figure 6A and B), suggesting knockdown of Sp1 may suppress the stem-cell like phenotype in CNE2 cells. Bmi1, c-Myc, KLF4, SOX2, NANOG and OCT4 are the key stem cell transcription factors (SCTFs). ABCG2, an ABC transporter member, is associated with the formation of side population and stem/progenitor features in nasopharyngeal carcinoma [38]. We therefore determined whether knockdown of Sp1-suppressed the stem-cell like phenotype was associated with changes in these molecules. Bmi1, c-Myc, KLF4 and ABCG2 were all decreased in Sp1-knockdown CNE2, HNE1 and HK1 cells (Figure 6C). In contrast, OCT4 and SOX2 was unaffected by Sp1 knockdown, indicating down-regulation of Sp1 suppressed the stem-cell like phenotype, depending on the reduced expression of Bmi1, c-Myc, KLF4, and ABCG2, but not OCT4 and SOX2 (data not shown).

Taken together, our data suggest that Sp1 was enriched in NPC cell lines as well as tissues and are highly correlated with the NPC progression. Down-regulation of Sp1 may induce the expression of p27 and p21, as well as the expressions of the critical SCTFs, resulting in the suppression of cell proliferation, clonogenicity, anchorage-independent growth and the stem-cell like phenotype in NPC cells. These findings demonstrated the potential use of Sp1 inhibitor in the clinical therapy of nasopharyngeal carcinoma.