Cellular immortalization and neoplastic transformation

Keywords: :

• telomere
• telomerase
• cellular transformation
• cancer
• karyotype
• chromosomal aberration
• oncogene
• tumorsuppressor gene
• microarray
• transcription profiling

A major difference between normal somatic cells and cancer cells is their proliferation potential. Normal somatic cells in culture are subjected to the well-known Hayflick limit; that is they stop proliferating after a limited number of divisions, while cancer cells can proliferate indefinitely. A large body of evidence has shown that telomeres, the terminal parts of the eukaryotic chromosomes, play a key role in defining cells’ proliferation potential. In fact, in the absence of telomerase, as in somatic cells, they shorten at each cell division and trigger cellular senescence when they reach a length below a threshold level.¹ In contrast, in tumors, induction of telomere maintenance mechanisms, prevalently telomerase activity, is a general finding, suggesting that progression of malignancy requires telomere stabilization.²

The seminal paper by Bodnar et al.³ demonstrated that restoration of telomerase activity allows several types of somatic cells to stabilize telomeres, bypass senescence and become immortal, thus defining a causal relationship between telomere maintenance, telomerase activity and cellular immortalization. Cellular immortalization per se, that is the increase of cells’ proliferative potential beyond the Hayflick limit, does not necessarily imply neoplastic transformation. Some authors have shown that telomerase immortalized cells can divide for many passages maintaining a normal phenotype, while others, our group among them, has demonstrated that the artificial extension of somatic cell's proliferative capacity can be associated with the development of the neoplastic phenotype (reviewed in ref. 4). In this second case, and in particular in the cellular system set up in our laboratory, named cen3tel, neoplastic transformation followed the lifespan extension achieved through telomerase expression and telomere stabilization. We could identify several phases along the road to transformation of cen3tel cells, each characterized by specific cellular and molecular features.^4-7

In a recent paper published in Cell Cycle, Duesberg and McCormack⁸ analyzed the karyotype of cen3tel cells at different propagation stages and concluded that immortality and tumorigenesis originated simultaneously in cells propagated for about 150 population doublings (PD) after exogenous telomerase expression, together with the acquisition of a “clonal and flexible karyotype” and independently of telomerase activity and telomere stabilization. Analyzing other cell lines expressing exogenous telomerase alone or telomerase and activated oncogenes, these authors reached the same conclusions; that is “clonal and flexible karyotypes” are at the basis of immortalization and transformation, while telomerase and oncogenes are not sufficient to trigger these processes.

Without questioning the role of karyotypic variations in tumorigenesis and the acquisition of a clonal and flexible karyotype by tumorigenic cen3tel cells, as described by Duesberg and McCormack,⁸ we would like to draw the attention on some published characteristics of the cen3tel cellular systems that undermine Duesberg and McCormack’s conclusions.

(1) The telomerase-negative cen3 primary fibroblasts, used as recipient to generate cen3tel cells, entered senescence around PD 40, stopped dividing and never became tumorigenic.⁵ Only cen3 cells expressing exogenous telomerase and proliferating beyond the Hayflick limit underwent neoplastic transformation. Thus, in this cellular system, telomerase expression was the prerequisite for the extension of the cells’ proliferative capacity and the development of the genomic variations driving tumorigenesis.

(2) In cen3tel cells, neoplastic transformation was a stepwise process; in fact, cells first became able to grow in agar and showed CDKN2A downregulation (around PD 100) then became tumorigenic.⁶ Cells around PD 100 had a transformed, even if not yet neoplastic, phenotype. Upon acquisition of the neoplastic phenotype, mutations in the tumor suppressor gene TP53 and overexpression of the c-myc oncogene were found in cen3tel cells. The same genes were found to be impaired in the independent cen3telS2 tumorigenic cells,^4,6 confirming the role of oncogenes and tumor suppressor genes in tumorigenesis. Moreover, genome-wide transcription analysis of cen3tel cells at different stages of transformation revealed that there was a sequential evolution of the transcriptome of these cells with the progressive modulation of more and more cancer-related genes, leading to the development of a more and more aggressive phenotype.⁷ Thus, it is hard to conclude that immortality and tumorigenicity originated simultaneously in cen3tel cells together with the clonal and flexible karyotype.

(3) Duesberg and McCormick⁸ stated that our results themselves indicate that immortalization and neoplastic transformation are independent of telomerase function. They report that according to our results, in cen3tel cells at PD 166, telomere length decreased and stabilized around values lower than in senescent cells. In Mondello et al.,⁵ we showed that cells around PD 100, and not around PD 166, had a mean telomere length lower than senescent cells; in addition, and more importantly, we pointed out that, despite the reduced length, cen3tel cell telomeres were functional. In fact, we did not detect telomeric fusions, and cells did not stop dividing, indicating that in the presence of telomerase, short telomeres can be stabilized and support cell growth. In a recent paper (Chiodi et al. Biochim Biophys Acta 2013; 1885-93), we have shown that in cen3tel cells just after PD 100, telomeres started to be elongated because of a change in telomere metabolism.

On the basis of the observations reported above, we think that cen3tel cells can be considered immortal all along their lifespan, since their generation through the introduction of the hTERT cDNA in cen3 primary fibroblasts. We do not see any experimental evidence that can postpone immortalization to the time of acquisition of the neoplastic phenotype and the occurrence of the “clonal and flexible karyotype.” Cellular immortalization can be viewed as a process that allows cells to keep on dividing; the selection of cells with karyotypic variations, a greater fitness and a tumorigenic phenotype can eventually occur, but this doesn’t necessarily imply that only these cells are immortal.

In conclusion, studies on the cen3tel cellular system and on other telomerase immortalized cells strongly support the hypothesis that exogenous telomerase expression stabilizes somatic cell telomeres, allowing cellular immortalization. In the absence of telomerase, as in telomerase-deficient mouse cells, other telomere maintenance mechanisms can stabilize telomeres, allowing immortalization and cellular transformation.⁹ We believe that the cytogenetic results reported by Duesberg and McCormack⁸ are very interesting, giving a picture of the evolution of the karyotype during cellular transformation and highlightening chromosome rearrangements that could possibly have a role in tumorigenesis. However, we do not think that these data disprove the link between telomerase, telomere stabilization and cellular immortality.