Susceptibility of Human Oral Squamous Cell Carcinoma (OSCC) H103 and H376 cell lines to Retroviral OSKM mediated reprogramming

Cell culture

Human Oral Squamous Cell Carcinoma cell lines (OSCC), H103 (STNMP Stage I) and H376 (STNMP StageIII), were obtained from Prof Ian Patterson, University Malaya, Kuala Lumpur, Malaysia. Cell lines were cultured in DMEM/F12 supplemented with 10% fetal bovine serum (FBS) (Gibco/Invitrogen, Grand Island, NY, USA) and Hydrocortisone (0.5 ug/ml) (Sigma-Aldrich, St. Louis, MO, USA) at 37 °C in the presence of 5% CO2. Reprogrammed OSCC cells were maintained in conditioned hESC medium consisting of DMEM/F12 (Gibco/Invitrogen, Grand Island, NY, USA), supplemented with 20% knock-out serum replacement (Gibco/Invitrogen, Grand Island, NY, USA), 0.1 mM nonessential amino acids (Gibco/Invitrogen, Grand Island, NY, USA), 4 mM L-glutamine (Sigma-Aldrich, St. Louis, MO, USA), 10 ng/mL fibroblast growth factor (bFGF) (Invitrogen, Carlsbad, CA, USA) and 0.1 mM 2-mecaptoethanol (Sigma-Aldrich, St. Louis, MO, USA). The medium was changed every 48 h.

Retrovirus production and infection

Retroviral vectors were produced via packaging cell lines, 293FT cell lines (human embryonal kidney cells) (Thermo Fisher Scientific, Waltham, MA, USA) using the Yamanaka Factors (Oct4, Sox2. Klf4, c-Myc). Vectors pMX-based retroviral hOct4 (Plasmid 17217) (Addgene, Cambridge, MA, USA), hSox2 (Plasmid 17218) (Addgene, Cambridge, MA, USA), hKlf4 (Plasmid 17219) (Addgene, Cambridge, MA, USA), hc-Myc (Plasmid 17220) (Addgene, Cambridge, MA, USA), retroviral gag–pol packaging plasmid (Plasmid 8449) (Addgene, Cambridge, MA, USA), VSV-G expression plasmid (Plasmid 8454) (Addgene, Cambridge, MA, USA) and pMX-GFP (Cell Biolabs, San Diego, CA, USA) used in these experiments were provided by Dr. Shigeki Sugii, DUKE-NUS Graduate Medical School, Singapore. Human embryonal kidney cell lines were plated at 3.6 × 10⁶ (70%–80% confluency) one day before transduction. The packaging cell lines were then transfected with Retro-GFP/OSKM vectors along with the transfection reagent LIPOFECTAMINE 2000 (Life Technologies, USA) according to the manufacturer’s protocol. The supernatant was collected at 48 h post-transfection and filtered with PVDF (45 uM) (Merck Milipore, Billerica, MA, USA). Polybrene (0.5 ul/ml from 10 mg/ml) was then added into the supernatant containing Retroviral Vectors to enhance the transduction efficiency within the target cell types. Presence of selection marker, green fluorescent protein (GFP) in H103 and H376 were counted manually using the cell count option in ImageJ program.

Retroviral infection and iPSCs cell generation

Human Oral Squamous Carcinoma Cells (H103 and H376) were seeded at 7.5 × 10⁴ cells/well on 6 well plates, 24 h before transduction. Equal amounts of supernatant containing each of the four retroviruses were collected and mixed prior to transduction of OSCC cell lines. Infected cells were incubated overnight. The medium was changed at 24 h post-infection to H103 and H376 specific medium. Transduced cancer cells were then subjected to spinfection at 1 h and 30 min (32 °C) before placing the plates into a hypoxic incubator (5% O₂). At day 3 post infection, OSCC cell lines were harvested by trypsinization and re-plated with 1 × 10⁴ cells/well on 6 well plates containing irradiated Mouse Embryonic Feeder Layer (MEF) (GlobalStem, Rockville, MD, USA). After 24 h, the medium was replaced with hESCs medium. Fifteen days after transduction, formed colonies were picked up and transferred onto a new mouse embryonic fibroblast (MEF) feeder layer (GlobalStem, Rockville, MD, USA). These colonies were propagated at least up to five passages before being subjected to pluripotency characterization.

Gene expression assessment in OSCC-induced pluripotent stem cells (iPSCs)

Ribonucleic acid (RNA) was extracted from parental H103 and H376 and respective reprogrammed counterparts using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocols. Total RNA was then converted into complementary deoxyribonucleic acid (cDNA) via reverse transcription (QuantiTech Reverse Transcriptase Kit).The fluorescent signals from each sample were plotted against cycle numbers which represented the accumulation of product over the duration of the quantitative real-time PCR (qRT-PCR) experiment. Quantitative real-time PCR amplification efficiency generated with primers (Table S1) were optimized using hESC, a commonly used gold standard for pluripotency characterization. Beta-actin (ACTB) was used as a house keeping gene in this experiment. Briefly, the condition used in this experiment comprises initial denaturation at 95 °C for 15 min, followed by 40 cycles of denaturation at 94 °C for 30 s with annealing at 60 °C for 30 s and extension at 72 °C. Following qRT-PCR, fluorescence data collection was performed during extension. Real-time PCR was performed with iQ5 Bio-Rad qPCR machine (Bio-Rad, Hercules, CA, USA) using Quantitect Sybr Green PCR Master Mix (Qiagen, Hilden, Germany). iQ5 Optical System Software, Version 2.0 was used for the analysis (Bio-Rad Laboratories, Hercules, CA, USA).

Immunofluorescense staining

Cells were fixed using 4% v/v paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA), washed three times with PBS containing 1% BSA and permeablized using Perm Buffer (BD Biosciences, San Jose, CA, USA) for 15 min at room temperature (Intracellular markers). After permeabilization, cells were blocked with PBS containing 1% BSA for 1 h at room temperature after the blocking solution was removed and cells were washed three times with PBS. Cells were then incubated with conjugated antibodies in PBS containing 1% BSA overnight at 4 °C. Antibodies used were at 1:50 dilution factor for Oct4-PE (BD Biosciences, San Jose, CA, USA), Sox2-PE (BD Biosciences, San Jose, CA, USA) Nanog-ALexa Fluor 488 (BD Biosciences, San Jose, CA, USA), and Tra-1-60-PE (BD Biosciences, San Jose, CA, USA). After the overnight incubation, cells were washed three times with PBS and stained with DAPI ANTIFADE GOLD (Invitrogen, Carlsbad, CA, USA) prior to viewing under Zeiss Imager A.1 Fluorescence Microscope (Carl Zeiss, Oberkochen, Germany).

Embryoid bodies (EB’s) formation

Rep-H103 cells were seeded onto ultra-low attachment plates (Corning) containing commercialize embryoid bodies medium (Millipore, Billerica, MA, USA). Transferred cells were grown in suspension for 8 days. The medium was consistently changed every 2–3 days up to 8 days without disrupting the EBs. EBs formed were carefully collected for immunofluorescence staining to determine the presence of three germ layers specific markers (Ectoderm: OTX2, Sox1; Endoderm: Sox17, Gata4; Mesoderm: Brachyury) according to the manufacturer’s instructions (Human Three Germ Layer 3-Color Immunostainining Kit) (R&D Systems, Inc., Minneapolis, MN, USA). Stained cells were observed under Zeiss Imager A.1 Fluorescence Microscope (Carl Zeiss, Oberkochen, Germany).

Directed differentiation of human H103 iPSCs like cells

Since H103 cell line is of the ectoderm and endoderm lineage (Jones & Klein, 2013), reprogrammed H103 was subjected to directed differentiation into adipocytes and osteocytes which is of the mesoderm lineages in order to access its differentiation potential.

Osteogenic assay

Osteogenic differentiation medium (ODM) (Erenpreisa & Cragg, 2013) was used to induce mineralization or osteogenesis in reprogrammed H103. IPSC-like cells were plated on gelatine coated six-well plates and cultured in 3ml of ODM consisting of 90% DMEM/F12 (Gibco/Invitrogen, Grand Island, NY, USA), 10% FBS (Gibco/Invitrogen, Grand Island, NY, USA) supplemented with 10nM dexamethasone (Sigma-Aldrich, St. Louis, MO, USA), 20 mM β-glycerol phosphate (Sigma-Aldrich, St. Louis, MO, USA), and 50 µM L-ascorbic acid (Sigma-Aldrich, St. Louis, MO, USA). The medium was changed every 2–3 days. After 21 days of incubation at 37 °C, 5% CO₂, cells were stained in Alizarin Red for visualization of calcium deposits. Stained cells were then evaluated under Eclipse TS100 inverted microscope (Nikon, Japan) and images were captured for analysis. Presence of mineralized osteoblasts indicated bright orange-red precipitate.

Adipogenic assay

Colonies plated on gelatine coated six-well plates and were cultured in 3ml of adipogenic differentiation medium (ADM) (Erenpreisa & Cragg, 2013) per well. ADM consisted of 90% DMEM/F12 (Gibco/Invitrogen, Grand Island, NY, USA), 10% FBS (Gibco/Invitrogen, Grand Island, NY, USA) supplemented with 0.5 µM dexamethasone, 0.5 µM isobuthylmethylxanthine and 50 µM indometacin. The medium was changed every 2–3 days. After 21 days of incubation at 37 °C, 5% CO₂, cells were stained in Oil Red O for visualization of lipid droplets. Stained cells were then viewed under Eclipse TS100 inverted microscope (Nikon, Japan) and images were captured for analysis. Mature adipocytes containing intracellular lipid vesicles were stained bright red.

Statistical analysis

Statistical data analysis was carried out with Paired t-Tests to compare the quantitative outcome of parental, reprogrammed counterparts at passage 5 and reprogrammed counterparts at passage 10 of H103 and H376 cell lines using SPSS, Software version 22.0 (IBM Corp, USA). All tests were conducted at 95% confidence level and all data were presented as mean  ± standard error of mean SEM. The differences were considered significant at P < 0.05.

Transfection efficiencies in oral squamous cells carcinoma cell lines

Green Fluorescence Protein (GFP) serves as an internal control and used for evaluating transduction efficiency. Uptake of vector pMX-GFP was tested using the highly transfectable human embryonic kidney cells (293FT) (Fig. S1). Subsequently, transduction efficiencies were examined in OSCC cell lines using vector pMX-GFP (16.5 µg) which encodes for green fluoresecent protein signals to confirm the uptake of transgenes prior to OSKM transduction. GFP signals were detected in H103 and H376 at 48 h confirming the uptake of transgenes (Fig. 1A). The transduction efficiency of 75.50% ± 1.52 and 38.20% ± 0.60 obtained from H376 and H103 cells respectively showed an almost 2-fold increase in H376 (Fig. 1B). Negative control was performed to access the potential influence of reprogramming on the viability of the cells (Fig. S2).

Reprogramming of oral squamous cell carcinoma cell lines

Clones were picked approximately two weeks after transduction (Fig. 2A) in both H103 and H376 cells. Distinct morphological patterns and changes were observed between the parental cancer cells and their reprogrammed counterparts (Fig. 2B i–iii and 2C i–iv). Only Rep-H103 clones were able to be passaged above passage 5 (Fig. 2B-iv) while clones derived from Rep-H376 differentiated at passage 2 onwards and were unable to sustain iPS-like cell morphology (Fig. 2C-v). Although higher GFP transfection capacity was achieved in H376, stable clones were successfully generated only from H103 cell line as these clones were able to be expanded up to passage 20 and still maintained an ESC-like morphology (Fig. 2B iv–v). Morphologies of all derived clones from H103 are highly distinct from its parental cells with clear borders on the mouse embryonic fibroblast feeder layer, high nucleus to cytoplasm ratio and the colonies displayed small cells morphology with spaces between them (Fig. 2B iv–v).

Differential gene expression

Overall, the expressions of endogenous Oct4, Sox2 in Rep-H103 cells were much higher than that of Rep-H376. However, suppression of Klf4 and c-Myc was seen in both reprogrammed cells (Fig. 3). Pluripotent marker expression levels of the reprogrammed cells relative to the parental counterparts which were quantified using qRT-PCR (Fig. 3A) showed down-regulation of Oct4 expression in rep-H103 at passage 5 but increased with a 2.80 fold change at passage 10. Expression of Sox2 (Fig. 3B) showed gradual up-regulation in Rep-H103 at passage 5 and passage 10 with 55 and77 fold change respectively. Oncogenic Klf4 (Fig. 3C) gene expression in reprogrammed H103 was significantly down regulated upon reprogramming and the level of expression maintained throughout passage 5 and passage 10. Similar gene expression pattern as Klf4 was observed in c-Myc gene (Fig. 3D) but at a much lower expression level. Nanog expression was also up-regulated gradually across passage 5 with 2.84 fold change to passage 10 with 7.07 fold change relative to its parental counterpart (Fig. 3E).

In the case of Rep-H376, Oct4 expression showed an initial reduction at passage 5 of −4.35 fold change but was up-regulated at passage 10 with 1.07 fold change (Fig. 3A). Gradual increase of Sox2expression (Fig. 3B) from passage 5 to passage 10 was observed in Rep-H376, a similar pattern to that of Rep-H103. Down-regulation of Klf4 expression was observed at passage 5 but slightly increased from −3.13 fold change to −1.89 respectively at passage 10 (Fig. 3C). Expression of c-Myc was almost not detected with −14.29 fold change at passage 5 and −100 at passage 10 (Fig. 3D). Nanog expression was found reduced in Rep-H376 at passage 5 with −1.79 fold change and showed little increase at passage 10 with 1.86 fold change (Fig. 3E).

Expression of pluripotency associated transcription factors

Rep-H103 continuously proliferated as adherent, flat colonies under feeder conditions and captured satisfactory pluripotent signals from real-time PCR analysis compared to reprogrammed H376. Therefore, successful Rep-H103 cells were characterised further for its pluripotency protein expression via immunofluorescence analysis. Rep-H103 expressed the pluripotency markers of Oct4, Sox2, Nanog and Tra-1-60 indicating a distinct difference between H103 derived iPS-like cells (Fig. 4A) over its parental cancer cell line (Fig. 4B).

Differentiation: embryoid body formation

Three dimensional (3D) sphere-shaped structures in suspension culture formed within a week in EB specific medium. At day 8, EBs were then evaluated for the presence of three germ layers via immunofluorescence staining. EBs was also induced from the parental cell line of H103 to serve as control (Fig. 5A-i). Rep-H103 showed morphologically compact round borders EB (Fig. 5A-ii). As H103 cell line was derived from both ectodermal and endodermal lineage expressions, signals from these two lineages were expected to be detected in EBs from both the parental and the reprogrammed counterpart (Figs. 5B-i and 5B-ii). However, mesoderm (Fig. 5B-iii) expression was only detected in EBs derived from Rep-H103 indicating the cross lineage differentiation ability of the Rep-H103 upon reprogramming into an iPS-like cell.

Directed differentiation assay

Crystal formation was observed during the osteogenesis process whereby the volume of the matrix mineralization increased during the 21 days of incubation as indicated from the Alizarin Red S stained calcium deposits (Fig. 6A-ii). In addition, tiny vesicles containing lipid droplets had formed in the cytoplasm of the cells after 21 days of incubation. The accumulation of lipid droplets was stained positive with Oil Red O staining (Fig. 6A-iv).