HeLa, SiHa, Ca Ski, and C-4 I human cervical cell lines were maintained in Dubelcco's modified Eagle's medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco), 100 units/ml penicillin (Gibco) and 100 μg/ml streptomycin (Gibco). In all experiments, cells were maintained at 37°C, 5% CO2, and 95% air atmosphere. All of our experiments were performed on cultures that were 70% confluent.
Culture and passage of tumor spheres
Cells were placed in serum-free medium (SFM) mixed with 20 ng/ml epidermal growth factor (EGF, Invitrogen), 20 ng/ml basic fibroblast growth factor (bFGF, Invitrogen), and 0.4% bovine serum albumin (BSA, Sigma-Aldrich) at a density of 1 × 103 cells/ml and cultured in ultra low attachment plates (Costar). The sphere culture media was changed every 48 h until the majority of spheres reached 100-400 μm in diameter (approximately 7 days). Spheres were collected by gentle centrifugation (5 min at 2,500 rpm), dissociated with EDTA 1.0 mM, and mechanically disrupted with a fire-polished Pasteur pipette. The cell suspension was sieved through a cell strainer with 40-μm nylon mesh to achieve a single-cell suspension and then re-plated in complete fresh medium.
Sphere-forming efficiency assay
For analysis of sphere formation, subconfluent cells were thoroughly dissociated with EDTA to prepare a single-cell suspension that was always visually verified. Then, 100 cells per well were plated in 96-well culture dishes in 200 μl of growth medium, and 25 μl of medium per well was added every 2 days. The number of spheres for each well was evaluated 7 days after seeding and sphere formation rate was counted. Sphere-forming efficiency (SFE) was calculated as the number of spheres formed divided by the original number of single cells seeded and expressed as a percentage . SFE was calculated from first through the fifth generation. All experiments were done in triplicate.
To examine cervical tumor-like epithelial differentiation of anchorage-independent cells, spheres were dissociated with EDTA and single cells were plated onto glass cover slips pre-coated with poly-L-lysine (Sigma-Aldrich) under standard conditions, DMEM/F12 supplemented with 10% FBS without growth factors. Cells were placed on cover slips, fed with FBS-supplemented medium every 2 days and processed 7 days after plating. Immunocytochemistry was performed with human anti-cytokeratin (AE1/3 clone, 1:50 final dilution, Dako) and visualized with LSAB + System-HRP (Dako). The chromogene was 3,3'-diaminobenzidine tetrahydrochlorate (Dako) solution developed with H2O2. Sections were counterstained with hematoxylin, dehydrated and cleared in xylene, and then mounted with Eukitt medium.
Each well (35-mm) of a six-well culture dish was coated with 2 ml bottom agar-medium mixture (DMEM, 10% FBS, 0.6% agar). After solidifying, 2 ml top agar-medium mixture (DMEM, 10% FBS, 0.3% Noble agar, BD Difco) containing cells was added, and dishes were incubated at 37°C for 21 days. Colonies were fixed with 6.0% glutaraldehyde, stained with 0.5% crystal violet, and counted using a dissecting microscope to determinate plating efficiency (PE).
Surface marker analysis by flow cytometry
One million cells were incubated for 10 min at 4°C with FcR blocking reagent (Miltenyi Biotech), followed by labeling with monoclonal antibodies to CD34 (AC136 clone, 1:11 dilution, Miltenyi Biotec), CD49f (GoH3 clone, 1:11 dilution, Becton Dickinson) and CD133 (AC133/1 clone, 1:11 dilution, Miltenyi Biotec). After 15 min at 4°C, cells were extensively washed with ice-cold PBS and resuspended in 0.5 μg/ml propidium iodide (PI) for 10 min at room temperature, and shielded from light prior to flow cytometry analysis with FACSCalibur (BD Biosciences, CA, USA) and Cell Quest Pro software (BD Biosciences, CA, USA). Fluorochrome-conjugated isotype-matched monoclonal antibodies from the same manufacturers were utilized to establish background fluorescence. CD133+ and CD34+ hematopoietic stem cells from umbilical cord blood were obtained from consenting mothers after full-term delivery and used as positive controls for CD133 and CD34 labeling. Human peripheral blood leukocytes were used as positive control for CD49f labeling.
Characterization of SP population
One millions of cells were resuspended in 1.0 ml of pre-warmed DMEM at 37°C containing 2% FBS, 1 mM Hepes, and 5 μg/ml Hoechst 33342 (Sigma-Aldrich) and were incubated under constant and slow agitation for 90 mn at 37°C. When Verapamil was used to block Hoechst efflux, cells were stained as described in the presence of 50 μM Verapamil (Sigma-Aldrich) . After staining, cells were washed with Hank's buffered salt solution, resuspended in ice-cold staining media, and then maintained on ice until their analysis or sorting. Immediately before analysis, PI was added at a final concentration of 2 mg/ml. Bone marrow cells were used as positive control. Murine bone marrow was extracted from the femurs and tibias of C57B1/6 mice, a single-cell suspension was made by passage of the bone marrow through an 18-gauge needle, and the cells were pelleted by centrifugation. Analysis and sorting were performed with a MoFlo Cell Sorter (Beckman Coulter, CA, USA) employing 60-mW multiline ultraviolet (UV) from a coherent I-90 laser. Hoechst blue was detected at 424/24 nm and Hoechst red at 590/30. The channels were separated by a 555LP dichromic beam-splitter. Sorting was carried out using a 100-lm nozzle and at 35-W sheath pressure, utilizing the purity-1 mode. Dye effluxing cells appear as a low fluorescing population termed the side population (SP).
Isolation by magnetic (MCS) and fluorescence cell sorting (FCS)
Human cervical cells lines were magnetically labeled and separated by double passage with 1 μl CD133 (CD133/1 clone, Miltenyi Biotec) and 1 μl CD34 microbeads (QBEND/10 clone, Miltenyi Biotec) per 1 × 106 cells, utilizing the Miltenyi Biotec CD34 and CD133 cell isolation kit. Indirect and positive separation was performed for CD49f using anti-phicoeritrin (PE) microbeads (PE4-14D10 clone, MiltenyiBiotec) and anti-CD49f-PE (GoH3 clone, Becton Dickinson). Magnetically enriched populations and SP were sorted with a MoFlo Cell Sorter (Beckman Coulter, CA, USA). At least 5,000 events were acquired for each sample and cells positive for PI were gated out. After separation by MACS and/or FACS, aliquots of positive and negative sorted populations were evaluated for purity by flow cytometry with a FACSCalibur (BD Biosciences, CA, USA) and Cell Quest software (BD Biosciences, CA, USA) using CD133/2-PE (239C3 clone, Miltenyi Biotec), CD49f (450-30A clone, Serotec), and CD34-PE antibodies (AC136 clone, Miltenyi Biotec). Purities ranged from 83 to 94% for positive and 89-99.7% for negative populations.
In vivo xenograft experiments
All animal experiments adhered to the requirements of NOM-062-ZOO-1999 Mexican Official Law and protocols were approved by the Ethics Committee of the Biomedical Research Institute, UNAM. Dissociated cells obtained under adherent and nonadherent conditions were counted and resuspended in 100 μl of FBS and growth factor-free DMEM at different cell densities. Cells were injected subcutaneously (s.c.) into the left and right flanks of 4-6-week-old female nude athymic mice (BALB/c-nu/nu). Engrafted mice were inspected weekly for tumor appearance by visual observation and palpation. Six weeks after transplantation mice were euthanized and tumor tissue collected. Tumors were digested usually during 1 h using dispase solution (1 mg/ml, Invitrogen) and 0.5% collagenase Type I (Invitrogen). They were stirred slowly at 37°C until the tissue was completely dissolved. Dispersed cells were then separated from residual tissue by passing the mixture through a 40-μm cell strainer to produce a single-cell suspension. Cells were obtained by centrifugation and plated in SFM. After reformation and dissociation of spheroids, single cells were reinjected into mice and, similar to the tumor mincing studies, the entire process was repeated. The remaining tumor, fixed in 10% buffered formalin and embedded in paraffin was sectioned (5-μm) on a rotary microtome and subsequently stained with hematoxylin and eosin (H&E) for histological evaluation and immunohistochemistry. Immunostaining was performed using tissue sections mounted on poly-L-lysine-coated slides and dried at 37°C overnight. The slides were deparaffinized in xylene and rehydrated. Endogenous peroxidase was blocked with 3% H2O2 for 5 min. To reduce nonspecific binding, the sections were incubated with 20% normal goat serum for 30 min at room temperature. Cells expressing human epithelial cytokeratins were identified after overnight incubation at 4°C with anti-cytokeratin AE1/3, as described above.
cDNA microarray analysis
Total RNA was extracted utilizing AllPrep DNA/RNA Micro Kit (Qiagen) and reverse-transcribed with the MessageAmp™ II aRNA Amplification Kit (Ambion) to generate Cy3-and Cy5-labeled (Ambion) cDNA probes for adherent and nonadherent samples, respectively. The labeled probes were hybridized to a cDNA Stanford University Microarray containing 40,000 immobilized oligonucleotide probes. Different fluorescently labeled cDNA probes were mixed in 45 μl of hybridization buffer DIG Easy Hyb (Roche), applied to the microarray and incubated at 37°C for 16 h. After hybridization slides were washed with 0.1 X SSC/0.1% SDS, 0.1 X SSC and 0.01 X SSC for 5 min at room temperature. Fluorescence intensities of Cy3 and Cy5 targets were measured and scanned separately using the GenePix 4100A Array Scanner (Axon Instruments, CA, USA). Images and quantitative data of gene-expression levels were processed and analyzed by GenePix Pro V5.0 (Axon Instruments, CA, USA). The array data for expression of 12,500 genes were filtered and genes with a median expression increased or decreased by a factor of at least 1.5-fold were selected. A total of 638 genes in HeLa and 857 genes in SiHa were clustered by the use of WebGestalt (Gene Set Analysis Toolkit). Two independent experiments were performed.
Cells from both monolayer cultures and 7-day-old spheres were enzymatically dissociated with EDTA and mechanically dissociated with a Pasteur pipette, both passed through a 40-μm sieve and immediately irradiated as single-cell DMEM suspension at a dose rate of 240 cGy/min for the time required to generate dose curves of 2, 4, 6, 8, and 10Gy with a linear accelerator Clinac 2100C, (Varian Medical Systems, CA, USA), operating at 6 MV. Corresponding controls were sham-irradiated. Colony-forming assays were performed immediately after irradiation by plating cells into triplicate six-well culture dishes. After 21 days, colonies were fixed with 6.0% glutaraldehyde, stained with 0.5% crystal violet, and counted. To generate a radiation survival curve, the surviving fraction (SF) at each radiation dose was normalized to that of the sham-irradiated control, and curves were fitted using a linear-quadratic model, SF. = exp(-αd - βd2) in which α is the number of logs of cells killed per gray from the linear portion of the survival curve, and β is the number of logs of cells killed per gray2 from the quadratic component. Three independent experiments were performed.
The approximation of data distribution to normality was preliminarily tested. Results are presented as mean and differences of the means with standard deviations (SDs) or 95% confidence intervals (CI). Statistical analysis was performed using factorial design analysis of variance (ANOVA) and Student two-tailed and unpaired t tests as appropriate utilizing Minitab 15™ software. Statistical significance when p < 0.05.