Spontaneous immortalization of neural crest‐derived corneal progenitor cells after chromosomal aberration

Abstract

Objectives:  In a previous study, we have reported the existence of neural crest‐derived stem cell‐like cells originating from the corneal limbus of juvenile mice (termed murine corneal cells, MCCs). To yield a sufficient number of MCCs, for example, for cell‐therapy approaches, here we have investigated MCCs’ ability for extensive proliferation, and we have evaluated their stem cell qualities and genetic stability after large‐scale culture.

Materials and methods:  MCCs were established from corneal limbal tissue of juvenile mice. To determine their cell proliferation and self‐renewing potential, MTT tests and an estimation of colony forming unit efficiency were carried out. Multipotency of cell differentiation was examined by applying adipogenic and osteogenic differentiation protocols. Moreover, karyotyping was performed and expression of stem cell markers and cell cycle‐associated genes was analysed.

Results:  MCCs, as primary cells, could be cultured for more than 60 passages. We observed increased cell proliferation and high number of colony forming units (CFUs) after extensive culture. Interestingly, there were no changes in expression of MCC markers. Furthermore, cell differentiation potentials remained comparable with MCCs at early passages. However, karyotyping revealed numeric chromosomal aberrations at higher passages. Moreover, tumour suppressor genes such as p16 and p21 were found to be down‐regulated after large‐scale cell culture.

Conclusions:  MCCs immortalize spontaneously after extensive cell culture, but still demonstrate stem cell‐like qualities.

Introduction

As the outermost surface of the eye, the cornea not only functions as a strong protective shield against external insults, but also provides optical function by transmitting and focusing light on the retina. It is comprised of three structurally unique and highly specialized cell layers – the epithelium, stroma and endothelium – whose homeostasis, deturgescence and integrity are essential for visual acuity. These layers are constituted of several different cell types including epithelia, endothelial cells and keratocytes, the last shaping the corneal stroma (1, 2, 3, 4).

Among these heterogeneous corneal cell populations, which all provide the cornea with its essential character, different types of stem or progenitor cells have been identified and the cornea has become of interest as an area of stem cell research (3, 4, 5, 6, 7, 8, 9). Stem cells have proven to be an intriguing source of tissues for use in cell‐based therapy and corneal tissue engineering. Keratoplasty, currently the only effective method of providing recovery of vision after corneal blindness, raises many problems and there is significant interest in alternatives. Stem cells are promising tools for new therapeutic strategies, but to properly exploit the recent advances of corneal stem cell research, many questions remain to be answered (4, 10).

In a previous study, we have identified a neural crest‐derived corneal stem‐like cell population (named murine corneal cells, MCCs). MCCs originate from the corneal limbus of juvenile mice before the stage of eyelid opening. They express a unique marker profile including typical neural crest‐originated stem cell transcripts such as Sca1 and other stem/progenitor cell markers such as Notch1, Nes, Abcg2 and Cd34. MCCs have further prove stem cell‐like attributes, such as cell migration, proliferation and, most interestingly, multipotency under in vitro conditions after differentiation into cells with features resembling adipocytes, osteoblasts and neuronal cells (11).

To use MCCs in cell‐based therapies of corneal tissue, high cell numbers are recommended. MCCs are somatic stem‐like cells and have a limited lifespan that generally restricts their value for cell‐based therapies. In our previous study, we observed a significant decrease in cell proliferation during the first cell passages although our investigations were conducted up to passage 10 (11). In the present study, we have focused on MCCs at later passages and investigated to what extent their sub‐culture would be possible and whether they maintain their stem cell‐like qualities and genetic stability after extensive proliferation.

Materials and methods

Isolation and cell culture

The MCCs were established from wild‐type C57Bl6/J mice purchased from Charles River Laboratories (Wilmington, MA, USA). All animals were housed and handled in full accordance with institutional guidelines and killed at maximum 8 days of age. Cells were isolated as described previously (11). In brief, discs of whole corneas including the limbal area of at least five eyes were chopped into fine pieces, rinsed and collected in PBS (PAA, Pasching, Austria). After being centrifuged at 800 g for 2 min, tissue pellets were resuspended in growth medium, high‐glucose DMEM (4.5 g/l of glucose) with l‐glutamine (PAA), supplemented with 10% foetal bovine serum (PAA) and 1x penicillin/ streptomycin (PAA). Tissue pieces were subsequently seeded in 25 cm² culture flasks (Nunc, Denmark); they attached to culture dish surface and after 3–5 days, an exodus of single cells could be observed. Cultures were maintained in humidified air (5% CO₂) at 37 °C and medium was changed twice a week. Upon reaching confluence, cells were detached using trypsin–EDTA (PAA) and again subcultured in growth medium, being plated at initial densities of 5–8 × 10³ cell/cm².

Colony‐forming unit assay

For measurement of colony‐forming units (CFUs), MCCs were plated in six‐well tissue culture plates (BD‐Falcon, Bedford, MA, USA) at initial density of 1 × 10³ cells/well and cultured in growth medium for 10 days. For better observation, CFUs appearing were stained with crystal violet ACS reagent (Sigma‐Aldrich, Taufkirchen, Germany). Colonies above 50 cells were counted using phase contrast microscopy (Nikon Eclipse TE 2000‐S; Nikon, Düsseldorf, Germany).

MTT assay

To investigate cell proliferation, MTT assays were performed. MCCs were seeded in six‐well tissue culture plates (Falcon) containing growth medium, at initial density of 5000 cells/cm² and were incubated for 72 h. Growth medium was then replaced with fresh medium containing 10% 3‐[4,5‐Dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazoliu‐mbromide (MTT) solution (Sigma‐Aldrich). After incubating for 4 h at 37 °C in 5% CO₂, dark purple formazon crystals formed were dissolved in 0.1 N HCl (Roth, Eggenstein‐Leopoldshafen Germany) in anhydrous isopropanol (Roth). This solution was subsequently measured photometrically at a wavelength of 540 nm (Ultrospec 2100pro; Amersham Biosciences, Buckinghamshire, UK).

Reverse transcription‐polymerase chain reaction

To analyse gene expression, total RNA was isolated from cultured cells using the NucleoSpin RNA II kit (Macherey‐Nagel, Düren Germany) including on‐column DNAse digestion according to manufacturer′s recommendation. RNA was quantified using a NanoDrop ND‐1000 Spectrophotometer (Thermo, Wilmington, DE, USA) and RNA quality was controlled by Agilent 2100 Bioanalyzer (Agilent, Wilmington, DE, USA). First‐strand cDNAs were synthesized according to instructions of the RevertAid H Minus First Strand cDNA Synthesis Kit (Fermentas, St. Leon‐Rot, Germany). PCR was performed using GoTaq Green Master Mix Kit (Promega, Mannheim, USA) at final volume of 20 μl and Thermocycler T3000 (Biometra, Göttingen Germany). Gene‐specific primers are provided in Table 1. PCR products were electrophoretically separated in 1.5% agarose gel. All genes were amplified for 35 cycles, whereas glyceraldehyde‐3‐phosphate dehydrogenase (Gapdh), which served as a housekeeping gene, was amplified for only 25 cycles.

Table 1.

 Gene‐specific primers for RT‐PCR

Gene	Primer sequence (5′‐3′)	Product size (bp)	GenBank accession ID
Notch1	Forward: TGCCTGTGCACACCATCCTGC	247	NM_008714
	Reverse: CAATCAGAGATGTTGGAATGC
Nes	Forward: AATGGGAGGATGGAGAATGGAC	496	NM_016701
	Reverse: TAGACAGGCAGGGCTAAGCAAG
Abcg2	Forward: CCATAGCCACAGGCCAAAGT	327	NM_011920
	Reverse: GGGCCACATGATTCTTCCAC
Musahi1	Forward: GGCTTCGTCACTTTCATGGACC	542	NM_008629
	Reverse: GGGAACTGGTAGGTGTAACCAG
Twist	Forward: CCAGAGAAGGAGAAAATGGACAGTC	259	NM_011658
	Reverse: AAAAAGTGGGGTGGGGGGACACAAA
Snail	Forward: CCCACTCGGATGTGAAGAGATACC	534	NM_011427
	Reverse: ATGTGTCCAGTAACCACCCTGCTG
Lum	Forward: TGCTGTCTCGGCTTCTCTGAAAG	567	NM_008524
	Reverse: AACATCCCCCACATTCCCAACC
Vim	Forward: ATGCTTCTCTGGCACGTCTT	206	NM_011701
	Reverse: AGCCACGCTTTCATACTGCT
Ki67	Forward: GAGCAGTTACAGGGAACCGAAG	262	X82786
	Reverse: CCTACTTTGGGTGAAGAGGCTG
Gapdh	Forward: GACCACAGTCCATGCCATCAC	453	NM_008084
	Reverse: TCCACCACCCTGTTGCTGTAG

Quantification of mRNA expression for cell cycle‐associated genes was performed by real‐time quantitative RT‐PCR (qRT‐PCR), using 7900HT Fast Real‐Time PCR System (Life Technologies, Applied Biosystems, Carlsbad, CA, USA). Gene‐specific primers are provided in Table 2. Relative differences in gene expression were calculated after normalization based on glucuronidase beta (Gusb) expression and were analysed using the ΔΔCT method (12).

Table 2.

 Gene‐specific primers for qRT‐PCR

Gene	Primer sequence (5′‐3′)	Probe IDa	GenBank accession ID
Tert	Forward: agagctttgggcagaagga	107	NM_009354
	Reverse: gagcatgctgaagagagtcttg
p16	Forward: cgacgggcatagcttcag	81	NM_001040654
	Reverse: gctctgctcttgggattgg
p21	Forward: tccacagcgatatccagaca	21	NM_007669
	Reverse: ggacatcaccaggattggac
Gusb	Forward: gtgggcattgtgctacctc	25	NM_010368
	Reverse: atttttgtcccggcgaac

^aProbe number from Roche Universal Probe Library.

Adipogenic and osteogenic differentiation

To initiate adipogenic and osteogenic differentiation in vitro, the MCCs were plated in six‐well tissue culture plates (BD‐Falcon) at initial density of 5000 cells/cm² and grown to subconfluence in normal growth medium. Subsequently, cells were switched to differentiation media and those left in growth medium were used as controls. Media were changed twice a week.

Adipogenic differentiation was performed in DMEM High glucose (4.5 g/l) with l‐glutamine (PAA), 10% FBS (PAA), 1x penicillin/streptomycin (PAA), 100 nm dexamethasone (Sigma‐Aldrich), 0.5 mm 3‐isobutyl‐1‐methylxanthine (Sigma‐Aldrich) and 10 μg/ml h‐insulin (Sigma‐Aldrich). Cell cultures were maintained for 14 days and finally analysed using oil red O staining.

For osteogenic differentiation, MCCs were cultured in DMEM high glucose (4.5 g/l) with l‐glutamine (PAA) containing 10% FBS (PAA), 1x penicillin/streptomycin (PAA), 0.1 μm dexamethasone (Sigma‐Aldrich), 100 μm l‐ascorbic acid 2‐phosphate (Sigma‐Aldrich), 2.9 mm KH₂PO₄ (Sigma‐Aldrich) and 20 mm Hepes (PAA). The experiment was conducted for 28 days and subsequently analysed using alizarin red staining.

Chromosome preparation

For numerical and structural chromosomal analyses of MCCs, karyotyping was performed. They were grown to subconfluence as a monolayer in culture flasks (Nunc, Roskilde, Denmark) in normal growth medium. Cells were then synchronized using thymidine solution (Sigma‐Aldrich) and subsequently treated with colcemid (Roche, Mannheim, Germany) for 120 min at 37 °C. After being detached from surfaces using trypsin–EDTA (PAA), cells were centrifuged. The cell pellet was resuspended and maintained in hypotonic solution (75 mm KCl) for 20 min at 37 °C, centrifuged again and finally fixed in methanol and acetic acid. Metaphase spreads were prepared on coverslips (Roth), dried overnight and Giemsa stained after trypsin pre‐treatment.

Results

MCCs surprisingly, could be grown for more than 60 passages and their proliferation levels increased even after passage 10 (Fig. 1). To estimate their self‐renewing potential at higher passages, CFU efficiencies were determined. Previous results revealed 22.3 ± 1.97 CFUs per 1000 cells in passage 2, which decreased significantly to 2.0 ± 0.58 CFUs per 1000 cells in passage 7 (11). New measurements using the same cell line demonstrate another significant increase to 62.2 ± 1.77 CFUs per 1000 cells in passage 57 (P < 10⁻¹⁰). Interestingly, no significant changes in morphology or size of colonies were observed during cell culture (data not shown).

Figure 1.

 Proliferation of MCCs in late passages. Proliferative capacity was measured at passages 15, 20 and 55. At passage 5, proliferation as determined in previous studies was set as 100%. Each bar represents average ± standard deviation of six biological replicates (*Student’s t‐test: P < 0.0000000001).

To examine eventual changes in gene expression profile of MCCs during cell culture, characteristic markers were compared at early and late passages (Fig. 2). These markers were determined in a previous study (11). Intriguingly, stem/progenitor cell markers such as Notch gene homolog 1 (Notch1), nestin (Nes), ATP‐binding cassette subfamily G member 2 (Abcg2) and Musashi homolog 1 (Musahi 1) were found at both passage 3 and passage 15. Neural crest markers twist gene homolog 1 (Twist) and snail homolog 1 (Snail) as well as markers associated with corneal stromal keratocytes such as lumican (Lum) and mesenchymal cell marker vimentin (Vim) were also detectable in both passage 3 and passage 15. Moreover, results indicate that expression of cell proliferation marker antigen, identified by monoclonal antibody Ki 67, (Ki67) might have been increased in passage 15.

Figure 2.

 RT‐PCR of MCCs at early and late passages. Total RNA was isolated at passage 3 and 15. RT‐PCR was performed with gene‐specific primers and analysed by agarose gel electrophoresis on a 1.5% gel. Gapdh was used as housekeeping gene. −RT = total RNA of MCCs without reverse transcriptase (negative control).

For further exploration of stem cell qualities of MCCs after extensive culture, the potential of cell differentiation was investigated by applying adipogenic and osteogenic differentiation protocols. Results show that these MCCs were capable of adipogenic and osteogenic differentiation, at both early and late passages (Fig. 3).

Figure 3.

 Adipogenic and osteogenic differentiation of MCCs at early and late passages. (a) MCCs at passage 4 revealed many lipid droplets (arrows) after being subjected to the adipogenic differentiation protocol. Cells were stained with oil red O. (b) MCCs at passage 24 also stained positively for lipid droplets (arrows). (c) At passage 4 cells revealed calcium accumulations (asterisk) after being subjected to the osteogenic differentiation protocol. These were stained with alizarin red. (d) Cells at passage 24 also stained positively for calcium accumulation (asterisk). Scale bars: 50 μm.

To evaluate genetic stability of the cells, chromosomal analyses were performed at both early and late passages. Karyotyping revealed that they had numerical and structurally normal karyograms of 40 chromosomes at passage 8 (Fig. 4). At passage 25, the cells no longer showed diploid chromosome complements but revealed huge numerical aberrations. Forty investigated metaphases demonstrated average of 87.8 ± 31.97 chromosomes per cell. At passage 55, 40 investigated metaphases demonstrated average of 70.6 ± 9.47 chromosomes per cell.

Figure 4.

 Karyotyping of MCCs at early and late passages. (a) MCCs at passage 8 revealed a numerical and structurally normal diploid karyogram of 40 chromosomes. (b) In contrast, at passage 25 they demonstrated huge numerical chromosomal aberrations. The given example displays 168 chromosomes found in a single cell. (c) Cells at passage 55 again revealed numerical chromosomal aberrations. The given example shows a nearly tetraploid complement of 76 chromosomes.

Expression of important cell cycle‐associated genes of early and late passages was analysed by qRT‐PCR measurements (Fig. 5). Interestingly, telomerase reverse transcriptase (Tert) was significantly down‐regulated after large‐scale proliferation. Moreover, tumour suppressor genes cyclin‐dependent kinase inhibitor 2A (p16) and cyclin‐dependent kinase inhibitor 1A (p21) also significantly decreased during the process of cell culture.

Figure 5.

 qRT‐PCR analysis of cell cycle‐associated genes in MCCs at early and late passages. Total RNA was isolated from the cells at passages 7, 15 and 55. qRT‐PCR was performed with gene‐specific primers for Tert, p16 and p21. Gusb was used as housekeeping gene. Each bar represents average ± standard deviation of three biological replicates (*Student’s t‐test: P < 0.005).

Discussion

Previously, we have reported isolation of neural crest‐derived cells originating from the corneal limbus of juvenile mice. These MCCs demonstrate stem cell‐like features, such as cell migration, proliferation, clonogenicity and multipotency, and they express stem or progenitor cell markers such as Sca1, Notch1 and Nes. Moreover, we have found that both cell proliferation and CFU efficiencies of the cells decline remarkably up to passage 10 (11).

In the present study, we have successfully subcultured MCCs beyond passage 10. They showed lifespan of at least 60 passages, which is unusual for primary somatic cells. Generally, the utility of primary cells is limited by their restricted lifetime (13). Nevertheless, for cell‐based therapy approaches of corneal tissue, MCCs should be present in high cell numbers. In addition, proliferation and CFU capacity demonstrated marked increase in the cells, outmatching passage 10, although applying identical cell culture techniques were used.

Intriguingly, MCCs upheld their characteristic gene expression profile, including stem/progenitor cell markers. Moreover, they maintained their multipotency during cell culture and even after late passages still differentiated into cells with features resembling adipocytes and osteoblasts.

However, our findings demonstrate that the MCCs acquired numerical chromosomal aberrations during cell culture, ranging from normal diploid chromosome number at early passages to more than tetraploid at later passages. It is known that aneuploidy is very frequently present in cells of solid tumours, and that tumour cells become increasingly aneuploid with tumour progression (14). Interestingly, decrease in chromosomes was observed between passages 25 and 55 for reasons unknown.

Additionally we analysed cell cycle‐associated genes, Tert, p16 and p21, which are often regulated in immortalized tumour cells. Results demonstrated that Tert was significantly down‐regulated in late passages. It is widely known that telomerase expression plays a substantial role in cell senescence, as it is normally repressed in postnatal somatic cells, resulting in progressive shortening of telomeres. Deregulation of telomerase expression in somatic cells may be involved in oncogenesis. Interestingly, studies in mice suggest that telomerase also participates in chromosomal repair (15, 16, 17). Furthermore, expression of p16 decreased as well in late passages. p16 is known to be an important tumour suppressor gene in mice (as in other species), and it frequently mutated or deleted in a wide variety of tumours. Increased expression of p16 usually reduces proliferation of stem cells and increases cell senescence (18, 19). However, we found the cell cycle inhibitor p21 to be down‐regulated. p21 plays a major role in DNA damage response and is also implicated in terminal differentiation and replicative senescence. It has additionally been shown that p21 knockout mice spontaneously develop tumours, which clearly shows the importance of p21 in tumour suppression (20, 21, 22). Considering these observations, we have drawn the conclusion that MCCs after extensive proliferation develop features resembling tumourigenic cells.

This study reports successful sub‐culture of MCCs up to at least until passage 60. At higher passages they retain expression of stem/progenitor cell markers and maintain stem cell‐like properties such as proliferation, clonogenicity and multipotency of differentiation. However, karyotyping of MCCs at higher passages revealed numerical chromosomal aberrations. Beside this, down‐regulation of tumour suppressor genes such as p16 and p21 underlined our assumption that these stem cell‐like cells underwent transformation and resembled tumourigenic cells during long‐term culture. This transformation might have taken place after passage 10 but before passage 15. Therefore, utility of MCCs for therapeutic applications might be limited and clearly restricted to those not older than passage 10, where their genetic stability has been assured. Nevertheless, establishing a novel, stable and well‐proliferating stem‐like corneal cell line, which maintains its multipotency of cell differentiation, should allow and facilitate further studies on corneal cell biology or stem cell biology.