Adult human mesenchymal stem cell as a target for neoplastic transformation

Abstract

The neoplastic process may involve a cancer stem cell. This concept has emerged largely from the careful analysis of tumour biopsy systems from haematological, breast and brain tumours. However, the experimental systems necessary to provide the cellular and molecular evidence to support this important concept have been lacking. We have used adult mesenchymal stem cells (hMSC) transduced with the telomerase hTERT gene to investigate the neoplastic potential of adult stem cells. The hTERT-transduced line, hMSC-TERT20 at population doubling level (PDL) 256 showed loss of contact inhibition, anchorage independence and formed tumours in 10/10 mice. hMSC-TERT4 showed loss of contact inhibition at PDL 95, but did not exhibit anchorage independence and did not form tumours in mice. Both lines had a normal karyotype but showed deletion of the Ink4a/ARF locus. At later passage, hMSC-TERT4 also acquired an activating mutation in KRAS. In hMSC-TERT20, expression of the cell cycle-associated gene, DBCCR1 was lost due to promoter hypermethylation. This epigenetic event correlated with acquisition of tumorigenicity. These data suggest that the adult hMSCs can be targets for neoplastic transformation and have implications for the development of novel anticancer therapeutics and for the use of hMSC in tissue engineering and transplantation protocols.

Main

Cancer is considered to be a disease of dysregulated cellular self-renewal capacity, yet a fundamental issue in cancer research is the identification of the actual cell type capable of sustaining the outgrowth of the neoplastic clone within solid tumours (Dick, 2003; Owens and Watt, 2003; Perez-Losada and Balmain, 2003). This capacity for self-renewal links neoplastic growth with stem cell biology (Reya et al., 2001; Preston et al., 2003). Stem cells within normal tissues are unique in their ability to both self-renew and to give rise to differentiated tissues. The identification of reservoirs of multipotential stem cells within adult tissue provides evidence that all tissues may have stem cells (Preston et al., 2003). Although tumour cells share characteristics in common with normal stem cells, the target cell for transformation has been elusive as it is not clear whether oncogenic transformation induces stem cell characteristics on normal lineage committed cells or whether an adult stem cell is indeed the target (Al-Hajj et al., 2003; Owens and Watt, 2003; Perez-Losada and Balmain, 2003). However, recent evidence from the study of human leukaemia shows that the normal haematopoietic stem cell and the neoplastic clone share common molecular mechanisms governing proliferation which is supportive of the normal haematopoietic stem cell being a target for transformation (Lessard and Sauvageau, 2003; Park et al., 2003). In addition, cancer stem cells have recently been isolated from breast and brain tumours (Al-Hajj et al., 2003; Hemmati et al., 2003; Passegue et al., 2003; Singh et al., 2003) The identification of the target cell for neoplastic transformation and the molecular mechanisms by which it subverts normal proliferative, cell fate and differentiation signals will be critical for the development of novel therapeutic strategies designed to specifically eradicate the cancer stem cell while sparing the normal stem cell population (Reya et al., 2001). Thus, the development of adult stem cell lines from a variety of tissues will be required to address the issue of whether human adult stem cells can be targets for transformation.

Adult human mesenchymal stem cells (hMSC) are notable for their ability to differentiate along multiple lineages including bone, cartilage, adipose and muscle cells (Jiang et al., 2002; Shi et al., 2002; Simonsen et al., 2002). Thus, the hMSC may be of use in the treatment of a diverse variety of clinical conditions (Collas and Hakelien, 2003; Korbling and Estrov, 2003). Furthermore, with the strong links between normal stem cells and cancer stem cells (Al-Hajj et al., 2003; Lessard and Sauvageau, 2003; Owens and Watt, 2003; Park et al., 2003; Perez-Losada and Balmain, 2003; Preston et al., 2003) the hMSC may be a particularly relevant model for investigating the neoplastic potential of adult stem cells.

We have recently developed cell line models immortalized with telomerase reverse transcriptase (hTERT) gene (Simonsen et al., 2002) for the hMSC. Ectopic expression of hTERT gene extends hMSC life and maintained the functional characteristics of the cells in spite of extensive proliferation. However, telomerase activity has been associated with neoplastic transformation in several models in vitro and in vivo (Wang et al., 2000; Hahn and Meyerson, 2001; Hamad et al., 2002).

We examined expression of transformed phenotype during long-term culture of three different independently growing cell lines (hMSC-TERT20, hMSC-TERT4, hMSC-TERT2) (Figure 1) derived from the parental cell strain. At various population doubling level (PDL), the cells were analysed for karyotype, doubling time (DT), labelling index (LI) using Ki67 staining, the presence of contact inhibition and anchorage dependence. Telomerase activity, levels of expression of hTERT transgene and the endogenous hTERT were determined using quantitative real-time PCR. Furthermore, the cells were mixed with MatrigelR and implanted subcutaneously in immune-deficient mice (NOD/LtSz-PrkdcSCID) for 6 months. HT1080 fibrosarcoma cell line was used as a positive control.

Figure 1
figure1

(a) Growth curves of three independent hMSC-TERT cell cultures. Yellow square hMSC-TERT2 was passage with split ratio 1:2; blue triangle hMSC-TERT4 was passage with split ratio 1:4; red circle hMSC-TERT20 was passage with split ratio 1:20. The population doubling levels abnormalities were first detected and indicated on the graphs for Ink4a/ARF locus deletion, KRAS oncogene mutation and DBCCR1 methylation. (b) Detection of deletion Ink4a/ARF (upper panel) and KRAS mutation (denaturing gradient gel electrophoresis; lower panel) in TERT4 cells. (c) Detection of DBCCR1 hypermethylation in TERT20 cells by methylation-specific PCR. DNA was treated with sodium bisulphite and PCR-amplified with primer pairs specific for unmethylated (U) and methylated (M) DBCCR1 alleles. SssI-methylated DNA provided a positive control for methylated DBCCR1 alleles. (d) Section of a tumour developed over 8 weeks after subcutaneous implantation of tumorigenic TERT cells mixed with MatrigelR in immune-deficient mice (NOD/LtSz-PrkdcSCID). The tumour shows high-cell density and invasive growth in mouse skeletal muscle stained with haematoxylin and eosin

The hTERT-transduced line was split into three independent hMSC-TERT cell cultures, namely, hMSC-TERT2 was passaged with split ratio 1:2; hMSC-TERT4 was passaged with split ratio 1:4; hMSC-TERT20 was passaged with split ratio 1:20. hMSC-TERT20 at PDL 256 showed loss of contact inhibition, anchorage independence and formed tumours in vivo in 10/10 mice (Table 1). Chromosomal analysis by G-Banding and comparative genomic hybridization (CGH) showed consistently normal 46, XY karyotype. The levels of hTERT transgene expression and telomerase activity did not differ between the cell lines and the endogenous hTERT expression was not detectable (not shown). Histologically, the tumours were mesenchymal and presented an expansive, nodular growth pattern. In some cases, invasive growth into skeletal muscle was observed. Immunohistochemically, the tumours were tested for antigens characteristic for different tissues. (Muscle: ASMA, pan actin, desmin, myogenin; bone: biglycan, osteonectin, bone sialoprotein [BSP],collagen type 1; Schwann cells: S100; Haematopoietic bone marrow cells: CD45; blood vessels: CD31, CD34; basement lamina formation: laminin, collagen 4; Epithelium: cytokeratin (CAM5,2) and additional markers: CD56, CD94). Moreover, Ki67/MIB1 used to examine the growth potential and demonstrated varies cell proliferation rates. The phenotypic profile indicated both smooth muscle and bone-forming cell properties. Organs from the animals with tumours did not present metastases. The hMSC-TERT4 showed loss of contact inhibition at PDL 168, but did not exhibit anchorage independence and did not form tumours in mice (Table 1). The DT of hMSC-TERT2 has remained low for 60 population doublings.

Table 1 Characteristics of human mesenchymal stem cells transduced with human telomerase reverse transcriptase gene (hMSC-TERT)

To characterize the molecular basis of the phenotypic changes in hTERT-transduced hMSC lines, we isolated DNA from different passages of hMSC-TERT4 and hMSC-TERT20 and searched for alterations of the BRAF, NRAS, KRAS, p16Ink4a and ARF genes. We first examined the integrity of the Ink4a/ARF locus by polymerase chain reaction (PCR) (Gronbaek et al., 2000) analysis of exon 2, which is common to p16Ink4a and ARF. PCR amplification yielded a band of the expected size for hMSC-TERT4 at PDL 74, whereas no product was obtained at PDL 95 or later passages (Figure 1b). Similarly, Ink4a/ARF exon 2 could not be amplified for hMSC-TERT20 from PDL 123 (data not shown). These data suggest that hMSC-TERT4 and hMSC-TERT20 have lost expression of the p16Ink4a and ARF tumour suppressor proteins due to homozygous deletion of the Ink4a/ARF locus whereas the Ink4a/ARF locus was detectable with FISH analyses suggesting that the deletion was a microdeletion. At a later passage, hMSC-TERT4 also acquired an activating mutation in KRAS, which may account for the observed increase in doubling time and loss of contact inhibition (Table 1; Figure 1b) To identify additional genetic changes responsible for the phenotypic changes in hMSC-TERT20, we compared the expression patterns of genes in early and later passages of this cell line using DNA microarrays. The cell cycle-associated gene, DBCCR1 was particularly interesting because its expression was entirely lost at later passages and while the exact function of this gene is elusive at present, it may be involved in regulating cellular senescence (Nishiyama et al., 2001). Methylation-specific PCR analysis showed that loss of DBCCR1 expression was due to hypermethylation of the DBCCR1 promoter and that this epigenetic event correlated with acquisition of the tumorigenic phenotype in hMSC-TERT20 (Table 1; Figure 1c).

The identification of the neoplastic potential within the normal adult human mesenchymal stem cell represents a major step forward in elucidating the origin and maintenance of tumours of mesenchymal origin such as soft-tissue sarcomas and osteosarcomas (Mackall et al., 2002; Helman and Meltzer, 2003). While these may be considered rare in comparison to many other solid tumours they have a very mixed clinical outcome and are currently poorly understood at the molecular level (Helman and Meltzer, 2003; Mackall et al., 2002). Thus, the availability of hMSC lines allows for some of the current issues relating to cell of origin and the molecular events that drive transformation to be addressed.

Studies on a variety of cell types derived from normal somatic tissues would suggest that ectopic expression of hTERT alone or in combination with inactivation of either the p53/p21CIP/WAF1 or the pRB/p16INK4a pathways will result in immortalization, without the acquisition of a tumorigenic phenotype (Kiyono et al., 1998; Hahn and Weinberg, 2002; Lundberg et al., 2002). Our results would at first appear at variance with these previous results. However, consequences of ectopic expression of hTERT may be different between adult somatic cells and stem cells. The adult stem cells are unique in its ability through asymmetric division to both self-renew and give rise to differentiated progeny. This delicate balance between proliferative capacity and differentiation is likely to be reliant on the precise regulation of many genes involved in cell cycle control, differentiation and immortality. While telomerase can supply the necessary mechanism to support long-term proliferation in lineage committed cell types, it is possible that stem cells require the alteration of pathways involved in differentiation or cell cycle blocks to efficiently achieve an extended proliferative lifespan and these may also predispose to neoplastic transformation. The recent finding that the Bmi-1 gene can determine the self-renewal capacity of adult haematopoietic stem cells (Lessard and Sauvageau, 2003; Park et al., 2003) would support the necessity for multiple alterations to achieve a true unrestricted proliferative lifespan and these may also predispose to neoplastic transformation. Consistent with our findings is the recent observation that ectopic expression of hTERT in the WI-38 fibroblast cell strain can also lead to the acquisition of additional genetic changes and a premalignant phenotype (Milyavsky et al., 2003). Thus it is possible that hTERT overexpression may favour tumorigenic events by affecting the function of cell survival genes, transcription factors and proliferation modulator genes such as p16INK4a through the induction of high cell proliferation.

In conclusion, we find that transducing hMSC with hTERT unmasks neoplastic potential and contributes to mesenchymal tumour development. Thus, the current study progresses our understanding of the molecular links between immortalization and transformation of stem cells. The identification of the target cell for neoplastic transformation and the molecular mechanisms by which it subverts normal proliferative and differentiation signals will be critical for the development of novel therapeutic strategies designed to specifically eradicate the cancer stem cell while sparing the normal stem cell population. In addition, the observed genetic and epigenetic changes and tumorigenic phenotype show that before the widespread use of stem cells can be used in the clinic, more complete understanding of the biological consequences of telomerase expression in adult stem cells is required.

References

  1. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ and Clarke MF . (2003). Proc. Natl. Acad. Sci. USA, 100, 3983–3988.

  2. Collas P and Hakelien AM . (2003). Trends Biotechnol., 21, 354–361.

  3. Dick JE . (2003). Proc. Natl. Acad. Sci. USA, 100, 3547–3549.

  4. Gronbaek K, de Nully Brown P, Moller MB, Nedergaard T, Ralfkiaer E, Moller P, Zeuthen J and Guldberg P . (2000). Leukemia, 14, 1727–1735.

  5. Hahn WC and Meyerson M . (2001). Ann Med., 33, 123–129.

  6. Hahn WC and Weinberg RA . (2002). N. Engl. J. Med., 347, 1593–1603.

  7. Hamad NM, Banik SS and Counter CM . (2002). Oncogene, 21, 7121–7125.

  8. Helman LJ and Meltzer P . (2003). Nat. Rev. Cancer, 3, 685–694.

  9. Hemmati HD, Nakano I, Lazareff JA, Masterman-Smith M, Geschwind DH, Bronner-Fraser M and Kornblum HI . (2003). Proc. Natl. Acad. Sci. USA, 100, 15178–15183.

  10. Jiang Y, Jahagirdar BN, Reinhardt RL, Schwartz RE, Keene CD, Ortiz-Gonzalez XR, Reyes M, Lenvik T, Lund T, Blackstad M, Du J, Aldrich S, Lisberg A, Low WC, Largaespada DA and Verfaillie CM . (2002). Nature, 418, 41–49.

  11. Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA and Klingelhutz AJ . (1998). Nature, 396, 84–88.

  12. Korbling M and Estrov Z . (2003). N. Engl. J. Med., 349, 570–582.

  13. Lessard J and Sauvageau G . (2003). Nature, 423, 255–260.

  14. Lundberg AS, Randell SH, Stewart SA, Elenbaas B, Hartwell KA, Brooks MW, Fleming MD, Olsen JC, Miller SW, Weinberg RA and Hahn WC . (2002). Oncogene, 21, 4577–4586.

  15. Mackall CL, Meltzer PS and Helman LJ . (2002). Cancer Cell, 2, 175–178.

  16. Milyavsky M, Shats I, Erez N, Tang X, Senderovich S, Meerson A, Tabach Y, Goldfinger N, Ginsberg D, Harris CC and Rotter V . (2003). Cancer Res., 63, 7147–7157.

  17. Nishiyama H, Gill JH, Pitt E, Kennedy W and Knowles MA . (2001). Oncogene, 20, 2956–2964.

  18. Owens DM and Watt FM . (2003). Nat. Rev. Cancer, 3, 444–451.

  19. Park IK, Qian D, Kiel M, Becker MW, Pihalja M, Weissman IL, Morrison SJ and Clarke MF . (2003). Nature, 423, 302–305.

  20. Passegue E, Jamieson CH, Ailles LE and Weissman IL . (2003). Proc. Natl. Acad. Sci. USA, 100 (Suppl 1), 11842–11849.

  21. Perez-Losada J and Balmain A . (2003). Nat. Rev. Cancer, 3, 434–443.

  22. Preston SL, Alison MR, Forbes SJ, Direkze NC, Poulsom R and Wright NA . (2003). Mol. Pathol., 56, 86–96.

  23. Reya T, Morrison SJ, Clarke MF and Weissman IL . (2001). Nature, 414, 105–111.

  24. Shi S, Gronthos S, Chen S, Reddi A, Counter CM, Robey PG and Wang CY . (2002). Nat. Biotechnol., 20, 587–591.

  25. Simonsen JL, Rosada C, Serakinci N, Justesen J, Stenderup K, Rattan SI, Jensen TG and Kassem M . (2002). Nat. Biotechnol., 20, 592–596.

  26. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J and Dirks PB . (2003). Cancer Res., 63, 5821–5828.

  27. Wang J, Hannon GJ and Beach DH . (2000). Nature, 405, 755–756.

Download references

Acknowledgements

We thank Drs Steen Kølvraa, Charlotte B Sørensen and Thomas J Corydon for critical readings of the manuscript. This study was supported by grants from Danish Medical Research Council, Danish Center for Stem Cell Research, the Novo Nordisk foundation and Danish Cancer Society.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Nedime Serakinci.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Serakinci, N., Guldberg, P., Burns, J. et al. Adult human mesenchymal stem cell as a target for neoplastic transformation. Oncogene 23, 5095–5098 (2004). https://doi.org/10.1038/sj.onc.1207651

Download citation

Keywords

  • mesenchymal stem cells
  • transformation
  • telomerase

Further reading