Human NB SH-SY5Y cells were obtained from the American Type Culture Collection (ATCC; Rockville, MD, USA). Cells were cultured in Dulbecco’s modified Eagle’s medium-F12 (DMEM-F12; Hyclone), with 10% fetal bovine serum (FBS; Gibco), 100 U/ml penicillin, and 100 μg/ml streptomycin (Sigma Chemical Co., St. Louis, MO, USA) and were grown in a 5% CO₂ incubator at 37°C. The TNKS1 inhibitor XAV939 was purchased from Sigma-Aldrich.

For side population (SP) method, 10⁶ cells were collected and incubated in pre-warmed DMEM-F12/2% FBS containing freshly added Hoechst 33342 (5 μg/ml final concentration) for 90 min at 37°C with intermittent mixing. In control group, cells were incubated with the Hoechst dye in the presence of verapamil (50 μM). At the end of incubation, cells were placed on ice at once and resuspended in cold DMEM-F12/2% FBS until fluorescence-activated cell sorting (FACS) analysis. Before analysis, cells were stained with PI (2 μg/ml). The Hoechst dye was excited with the UV laser at 351–364 nm and its fluorescence measured with a 515 nm SP filter (Hoechst blue) and a 608 EFLP optical filter (Hoechst red). The emission wavelengths were separated by a 540 DSP filter. For surface marker method, 107 SH-SY5Y cells were collected and suspended in FcR blocking reagent (Miltenyi Biotec) with 0.5% FBS in PBS and added with mouse anti-human CD133/2(293C3)-PE monoclonal antibody which was diluted at 1:11. In addition, mixed well and refrigerated for 10 min in the dark (4–8°C). Then cells were washed by adding 1–2 ml buffer per 10⁷ cells and centrifuged at 300 × g for 10 min, supernatant was aspirated completely and the cell pellet resuspended in DMEM-F12 with 2% FBS for isolating CD133⁺ and CD133⁻ cells, respectively, by flow cytometry. CD133⁺ cells were considered as NB CSCs. Mouse IgG2b-phycoerythrin was used as an isotype control and was set up for each sample.

The SH-SY5Y cells were plated at a density of 10⁵/ml in 6-well plates. Etoposide (V 4629, Sigma-Aldrich, dissolved in PBS) of 20, 40 and 60 μM were added two days after cell seeding respectively, maintained for 24 h, and removed by substituting the culture medium with fresh medium without drug. Untreated (control) cells were also subjected to change of medium at the same incubation times. Then the ratios of CSCs were analyzed by flow cytometry and the group of highest ratio was chosen. CSCs were sorted and cultured with DMEM/F12 medium contained 40 ng/ml bFGF, 20 ng/ml EGF as well as 2% B27. This kind of medium contributes to prevent the differentiation of CSCs and maintain its characteristic.

The isolated CD133⁺ and CD133⁻ cells were cultured with different condition respectively. Some of them were cultured on coverslip for the observation by scanning electron microscope (SEM). When cells were at 60–70% confluence, the culture medium was discarded and cells were fixed in cold mixture of osmium tetroxide and glutaraldehyde for 1–3 h. Then samples were dehydrated in graded ethanol series, replaced with isoamyl acetate, dried at critical point, coated with gold on surface, and observed by JSM-T300 SEM. Another portion of cells were cultured in 6-well plates. The monolayer cells were collected and fixed with 2.5% glutaraldehyde for 2 h. The cells were fixed with 1% osmium tetroxide for 2–3 h, dehydrated with graded ethanol, replaced with acetone, pelleted and embedded in 3% agar. The agarized pellet was cut into 1-mm slices and immersed in 3% uranyl acetate-lead citrate for double staining, and viewed with a transmission electron microscope (TEM, JEM1200EX).

The isolated NB CSCs were cultured and proliferated for XAV939 treatment and RNAi. XAV939 was dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich) and the final concentration was 1 μM based on our previous study results (16). The control group was treated with DMSO only. The TNKS1 gene was knocked down by TNKS1-shRNA (shRNA group) while cells in control group were transfected with lentivirus-mediated scrambled-shRNA (SCR group). The XAV939 treatment, viral transfection and control groups were maintained for 72 h. Then all groups were subjected to quantitative real-time RT-PCR (qRT-PCR) analysis, western blot analysis, immunofluorescence and migration assay.

Total RNA was extracted using the RNeasy Mini kit (Qiagen, Valencia, CA, USA) and quantitated using NanoDrop 1000 (NanoDrop, Wilmington, DE, USA). Then the RNA was reverse transcribed into cDNA using PrimeScript^® RT reagent kit with gDNA Eraser (Takara). QRT-PCR was performed using SYBR^® Premix Ex Taq™ II (Takara) on ABI 7500 Real-Time PCR system (Applied Biosystems). Sequences of the primers for CD133 were 5′-AGT GGCATCGTGCAAACCTG 3 (forward) and 5′-CTCCGAAT CCATTCGACGATA-3′ (reverse). Sequences of the primers for β-actin were 5′-TGGCACCCAGCACAATGAA-3′ (forward) and 5′-CTAAGTCATAGTCCGCCTAGAAGCA-3′ (reverse). The PCR amplification was done at: 95°C for 30 sec; 95°C for 5 sec, 60°C for 34 sec for 40 cycles. The relative changes in gene expression data were analyzed by the 2-ΔΔCT method. β-actin was used as an internal control. Triplicates were run for each sample in three independent experiments.

The NB CSCs were lysed with RIPA buffer and protein concentration was determined by the Bradford method. Equal amounts of protein (40 μg) were used for western blot analysis with primary antibodies to anti-CD133 (Bioss, bs-0209R) and anti-β-actin (Santa Cruz, sc-1616-R). Specific antibody binding was detected by horse-radish peroxidase-conjugated goat anti-rabbit antibodies and visualized with ECL reagent (Santa Cruz) according to the manufacturer’s protocol. Antibody to β-actin was used to evaluate protein loading in each lane.

The NB CSCs were fixed with 4% paraformaldehyde for 10 min at room temperature, and permeabilized with 0.5% Triton X-100 for 15 min. After blocking with 1% BSA for 30 min, cells were incubated with the primary antibody (CD133, Bioss, 1:200) for 2 h at 37°C, then incubated with secondary fluorescein conjugated antibody (Bioss, 1:500) for 1 h at 37°C in the dark. To ensure specificity of results, negative controls with no primary antibody or no secondary antibody were used. The coverslips were incubated with Hoechst 33342 (5 μg/ml for 10 min) for nuclear counterstaining, mounted with 95% glycerol (Sigma-Aldrich), and observed by fluorescence microscope (Olympus IX71). Five random fields were selected and positive cells were counted.

To evaluate migratory properties of NB CSCs treated with XAV939 or RNAi, migration assay was used. Cells (1×10⁵) were suspended in 100 μl culture medium without FBS and loaded onto the top of Transwell chambers (24-well plate) equipped with 8.0 μm pore-size polycarbonate membranes (Corning). Culture medium supplemented with 10% FBS was used as chemotactic stimuli in the bottom chambers. After 24 h of incubation, cells on the upper surface of the filter were mechanically removed with a cotton swab, and those which migrated underneath the surface were fixed with 90% ethanol and stained with 0.1% crystal violet dye. Photographs were taken using an inverted phase contrast microscope (Olympus CK40) and the number of cells was counted.

The results obtained from three independent experiments performed in triplicate are presented as mean ± SD. One-way ANOVA was performed for comparison between each group. P-values of <0.05 were considered as statistically significant.

The SP and non-SP (NSP) cells were sorted from the SH-SY5Y cell line. The P1 gate containing the SP cells accounted for 0.9±0.06% of the total cells (Fig. 1A). The percentage of SP cells decreased to 0.1% after treatment with verapamil, indicating that this population consisted of SP cells (Fig. 1B). However, the SP cell population is not obvious in flow chart and with poor specificity. Moreover, Hoechst 33342 is harmful to cells. So the SP cells after sorting had high death rate and poor culture result. SH-SY5Y cells were also analyzed by FACS using the characterization of CD133 expression on NB CSCs surface. The results showed that the ratio of control IgG group was 1.39% (Fig. 1C), while that of CD133 antibody group was 2.17% (Fig. 1D). Therefore the ratio of CD133⁺ cells in SH-SY5Y cell line was 0.78%.

CD133 are commonly used as cell surface markers representing the CSC subpopulation in NB (18). CSCs are resistant to conventional chemotherapy, and therefore we further investigate the enrichment of NB CSCs in SH-SY5Y cells using different concentrations of etoposide and sorted NB CSCs with the cell surface marker of CD133 (Fig. 2). After treatment with etoposide of 20, 40 and 60 μM, the flow cytometry analysis showed that the ratio of NB CSCs was 0.72, 0.49 and 1.37%, respectively. The ratio of NB CSCs in 60 μM etoposide treatment group increased to nearly 2-fold of non-treatment group (Figs. 1 and 2), and was the highest among all groups.

Under a SEM, NB CSCs were rounded or oval with rich and trivial cell processes on the surface, which gathered into a ball growth, indicating that it was in an undifferentiated state (Fig. 3A). However, the non-NB CSCs were elongated and spindle-like with poor but obvious cell processes (Fig. 3B). TEM showed the nuclear-cytoplasmic ratio of NB CSCs was higher than that of non-NB CSCs, and the NB CSCs had fewer cytoplasmic organelles than the non-NB CSCs (Fig. 3C and D). The undeveloped cytoplasmic organelles, such as rough endoplasmic reticulum or Golgi apparatus, indicated that the NB CSCs were in immature or juvenile stage.

QRT-PCR and western blot analysis were used respectively to validate the stemness change of NB CSCs after XAV939 treatment or RNAi-TNKS1. The results of qRT-PCR showed that the melting curve of β-actin and CD133 was a standard single peak, indicating specific amplification products (Fig. 4A). After XAV939 treatment or RNAi-TNKS1, the relative expression of CD133 mRNA was 0.57±0.09 and 0.52±0.11, respectively, which were both lower than those in control and SCR group (Fig. 4B, P<0.05). The relative expression of CD133 protein was significantly lower in XAV939 treatment group and RNAi-TNKS1 group compared with control groups (Fig. 4C and D, P<0.05). The results indicated that both XAV939 treatment and RNAi-TNKS1 repressed the stemness of CSCs to some extent.

The immunofluorescence results showed that CD133 protein was expressed in the cytoplasm and membrane of NB CSCs, and presented red fluorescence, while the nucleus of cells presented blue fluorescence (Fig. 5A). After XAV939 treatment or RNAi-TNKS1, the number of NB CSCs reduced and some signs of apoptosis appeared, including the condensation and bright stained of nuclear chromatin. The fluorescence intensity were analyzed with image software and presented in a histogram (Fig. 5B). The mean fluorescence intensity of CD133 protein in control, XAV939 treatment, SCR and shRNA group was 1±0.10, 0.21±0.06, 0.97±0.09 and 0.41±0.08, respectively. Compared with the respective control groups, XAV939 treatment and RNAi-TNKS1 significantly decreased the expression of CD133 protein (P<0.01). The change of mean fluorescence intensity of Hoechst 33342 showed similar tendency to that of CD133. The results indicated that XAV939 treatment and RNAi-TNKS1 reduced the expression of stemness of NB CSCs.

In this study, 9 high power field of view were selected randomly, and cells migrated to the bottom of the wells were counted. As shown in Fig. 6A, the mean number of the migrated cells in control, XAV939 treatment, SCR and shRNA group was 62.0±6.0, 36.7±4.5, 63.1±2.0 and 40.1±1.9, respectively (Fig. 6B, P<0.05). The results demonstrated that XAV939 treatment and RNAi-TNKS1 both decreased the migration of NB CSCs.