Defined genetic events associated with the spontaneous in vitro transformation of ElA/Ras-expressing human IMR90 fibroblasts.

In contrast to rodent cells, normal human fibroblasts are generally resistant to neoplastic transformation in vitro. Here, we report the derivation and characterization of a spontaneously transformed cell line from normal human IMR90 fibroblasts transduced with E1A and Ras oncogenes. Unlike the parental, non-tumorigenic E1A/Ras-expressing IMR90 cells, these spontaneously transformed cells displayed aberrant growth potential in vitro and were capable of tumorigenesis in vivo. In contrast to the parental E1A/Ras-expressing cells, both the spontaneously transformed cells and cells derived from resultant tumors displayed specific t(7q;8q) and t(5q;17) structural chromosomal changes. Chromosome 8q contains c-Myc, which is capable of activating the telomerase catalytic subunit hTERT. Notably, upregulation of c-Myc, hTERT and telomerase activity were detected only in the tumorigenic cells. Transduction of Myc siRNA into the tumorigenic cells led to a concomitant downregulation of hTERT. Furthermore, transduction of Myc or hTERT into the non-tumorigenic E1A/Ras-expressing IMR90 cells was able to confer tumorigenesis on these cells. These studies suggest that the t(7;8) translocation may result in Myc overexpression and its subsequent activation of hTERT, which may contribute to the tumorigenicity of the IMR90 cells. Furthermore, this report describes additional successful neoplastic transformation of human IMR90 fibroblasts by defined genetic elements. The spontaneously transformed cells we have derived provide a valuable model system for the study of neoplastic transformation.

Expression of p16INK4A variants in senescent human fibroblasts independent of protein phosphorylation.

Upregulation of the p16 tumor suppressor is a hallmark of senescence in human fibroblasts. In this study, we investigated potential protein modification of p16 in senescent human fibroblasts using 2D SDS-PAGE analysis. Three distinct p16 variants with isoelectric points of 5.2, 5.4, and 5.6, were consistently detected in normal human IMR90 fibroblasts that had undergone senescence due to forced expression of oncogenic H-ras or culture passage. Moreover, in contrast to short-term serum starvation, which induces quiescence, IMR90 fibroblasts cultured in low serum for a prolonged period exhibited senescent phenotypes and expression of the three p16 variants. All three p16 variants are unlikely phosphoproteins since they failed to react with antibodies against phospho-serine, and were resistant to the treatment with phosphatases. Functionally, co-immunoprecipitation assays using antibodies against cdk4 and/or cdk6 revealed that only the two most acidic p16 variants associated with cdk4/6. Moreover, senescence induced by the forced expression of p16 in early passage IMR90 fibroblasts or osteosarcoma U2OS cells was accompanied by expression of the two most acidic p16 variants, which also associated with cdk4/6. In summary, we report that prolonged serum starvation-induced senescence may provide an additional model for studying biochemical changes in senescence, including p16 regulation. Furthermore, induction of endogenous p16 in senescent human fibroblasts correlates with the expression of three distinct p16 variants independent of protein phosphorylation. Lastly, expression of the two cdk-bound variants is sufficient to induce senescence in human cells.

Molecular signature of oncogenic ras-induced senescence.

Senescence irreversibly arrests the proliferation of cells that have sustained significant cellular stress. Replicative senescence, due to the shortening and dysfunction of telomeres, appears to provide a barrier to the immortalization of cells and development of cancer. In normal human fibroblasts, senescence induced by oncogenic H-ras displays a nearly identical cellular phenotype to that of replicative senescence, suggesting the activation of a common senescence mechanism. In this study, we investigated the gene expression profile of oncogenic H-ras-induced senescent human diploid fibroblasts. We found altered gene expression of various cell cycle regulators in both oncogenic H-ras-induced senescent cells and replicative senescent cells. Similar to replicative senescent cells, H-ras-induced senescent cells exhibited specific downregulation of genes involved in G2/M checkpoint control and contained tetraploid cells that were arrested in a G1 state. This observation suggests that the inactivation of G2/M checkpoints may be involved in senescence and may play a role in the generation of senescent G1 tetraploid cells. Lastly, we have identified two genes, topoisomerase IIalpha and HDAC9, whose expression was specifically altered under several conditions associated with senescence, suggesting that these two molecules may be novel biomarkers for senescent human fibroblasts.

Differential oncogenic Ras signaling and senescence in tumor cells.

Several studies have shown that forced expression of oncogenic H-ras can induce a senescence-like permanent growth arrest in normal cells. Here we report that expression of oncogenic H-ras in human osteosarcoma U2OS cells also resulted in a senescence-like flat and enlarged cell morphology and permanent growth arrest. In contrast to normal human fibroblasts, U2OS cells were arrested independently of the p16 and ARF tumor suppressors. Treatment with a MEK inhibitor or a p38MAPK inhibitor interrupted oncogenic H-ras-induced growth arrest in U2OS cells, suggesting that activation of MAPK pathways is important. To further determine whether this process is unique to oncogenic H-ras signaling, we examined the effect of oncogenic K-ras on normal cells and human osteosarcoma cells. Similar to oncogenic H-ras, oncogenic K-ras also induced senescence in normal fibroblasts, while transforming immortalized mouse fibroblasts. However, in contrast to oncogenic H-ras, oncogenic K-ras failed to induce a permanent growth arrest in osteosarcoma U2OS cells. Additionally, cells transduced with oncogenic K-ras exhibited distinguishable cellular changes compared to those transduced with oncogenic H-ras. In summary, we report for the first time that oncogenic H-ras signaling can trigger a senescence-like growth arrest in tumor cells, independent of the p16 and ARF tumor suppressors. This result suggests that tumor cells may harbor a senescence-like program that can be activated by ras signaling. Moreover, our study uncovered a cell type-dependent differential response to oncogenic K-ras, as compared to oncogenic H-ras.