Unlike mouse cells, primary human cells are refractory to oncogenic transformation — transformation requires a specific combination of three genetic elements (the HRAS -V12 oncogene, the SV40 early region and the catalytic subunit of telomerase ( TERT )) as opposed to a combination of two oncogenes in mouse cells. But what characteristic might explain this difference? The relative ease of immortalization of mouse cells — because they have longer telomeres and express telomerase — is one possibility, and a requirement for telomere maintenance and immortality in human cells is supported by the fact that TERT is the third element. However, in the November issue of Cancer Cell, Yvette Seger et al. investigate this premise, and show that TERT does not have to be one of the three elements.

Expression of HRAS-V12 alone causes irreversible growth arrest, and adenovirus E1A is one of the few oncogenes that can rescue this phenotype; in fact, expression of HRAS-V12 and E1A is sufficient for transformation of mouse cells, so the authors investigated whether this oncogenic combination could also transform primary human fibroblasts — BJ cells. They first investigated whether cells expressing HRAS-V12 and E1A exhibited anchorage-independent growth — a characteristic of transformation — and found that they did. Co-expression of HRAS-V12 with E1A deletion mutants confirmed that E1A must maintain the ability to interact with p300, p400 and the retinoblastoma (RB) family. Interestingly, despite exhibiting characteristics of transformation, these cells are not able to form tumours when injected into immunocompromised mice. So, what other element might be required for this function?

The SV40 early region is known to abrogate both the RB and p53 pathways, so the authors investigated whether expression of the oncogene MDM2 , which inhibits p53, could confer tumorigenic potential on the HRAS-V12- and E1A-expressing BJ cells. Triple-infected cells (BJ/ERM cells) were injected into immunocompromised mice and were able to generate tumours with a similar latency to human cancer cell lines.

So, can transformation really occur in the absence of telomerase activity or an alternative telomere-maintenance strategy? Telomerase activity could not be detected using the TRAP assay in the BJ/ERM cells, and they also do not seem to be immortal — they undergo 'crisis' and adopt a senescent phenotype after prolonged culture. Similarly, the tumours that are formed from these cells do not generally express TERT, as shown by reverse-transcriptase polymerase chain reaction, and do not have telomerase activity, as shown by the TRAP assay. On explantation into culture, BJ/ERM tumour cells undergo crisis, which is indicative of a lack of telomere maintenance, and telomeric fluorescence in situ hybridization revealed that the telomeres continued to be eroded during tumour growth, confirming that telomeres were not maintained by the alternative (ALT) recombination-based mechanism.

Karyotypic analysis of chromosomes from explanted BJ/ERM tumour cells reveals many chromosomal abnormalities, which are characteristic of the end-to-end chromosome fusions that occur as telomeres shorten. This type of chromosomal instability could accelerate the tumorigenic process.

So, unlike previous transformation protocols, this one does not require telomerase activity or immortalization, demonstrating that immortality is not an obligate characteristic of a cancer cell.