To facilitate the study of human endothelial cells we have used a replication defective retrovirus encoding the catalytic subunit of telomerase (hTERT) to derive populations of telomerase-immortalized human microvascular endothelial (TIME) cells. Whereas parental HMVECs became senescent on average within 35-45 population doublings (PDs), TIME cells have continued to proliferate for at least 200 PDs. TIME cells express readily detectable telomerase activity but display only a modest increase in telomere length. Karyotypic analysis reveals the cells to have a normal complement of human chromosomes with no evidence of gross genetic abnormalities. Furthermore, TIME cells retain many of the characteristics of the primary endothelial cells from which they were derived. For example, they express a panel of characteristic endothelial cell surface marker proteins such as CD31/PECAM-1 and alpha(v)beta3-integrin. In addition, TIME cells express receptors for low-density lipoprotein (LDL) receptor as they are competent for receptor-mediated endocytosis of fluorescent acetylated LDL. Importantly, when plated on matrigel, TIME cells undergo tubule formation. Moreover, when cocultured in the presence of human glioma cells, but not primary human astrocytes, TIME cells are induced to form stable tubules. Detachment of TIME cells from extracellular matrix leads to a form of programmed cell death known as anoikis. Conditional activation of the protein kinase Akt (Akt:ER*) significantly inhibited the onset of TIME cell anoikis under these conditions. We believe that the ability of hTERT to immortalize primary human endothelial cells, and the fact that such cells retain the endothelial characteristics of the cells from which they were derived, will greatly facilitate the analysis of human endothelial cell biology in vitro.

Senescence and genomic integrity are thought to be important barriers in the development of malignant lesions. Human fibroblasts undergo a limited number of cell divisions before entering an irreversible arrest, called senescence. Here we show that human mammary epithelial cells (HMECs) do not conform to this paradigm of senescence. In contrast to fibroblasts, HMECs exhibit an initial growth phase that is followed by a transient growth plateau (termed selection or M0; refs 3-5), from which proliferative cells emerge to undergo further population doublings (approximately 20-70), before entering a second growth plateau (previously termed senescence or M1; refs 4-6). We find that the first growth plateau exhibits characteristics of senescence but is not an insurmountable barrier to further growth. HMECs emerge from senescence, exhibit eroding telomeric sequences and ultimately enter telomere-based crisis to generate the types of chromosomal abnormalities seen in the earliest lesions of breast cancer. Growth past senescent barriers may be a pivotal event in the earliest steps of carcinogenesis, providing many genetic changes that predicate oncogenic evolution. The differences between epithelial cells and fibroblasts provide new insights into the mechanistic basis of neoplastic transformation.

The genomic changes that foster cancer can be either genetic or epigenetic in nature. Early studies focused on genetic changes and how mutational events contribute to changes in gene expression. These point mutations, deletions and amplifications are known to activate oncogenes and inactivate tumor suppressor genes. More recently, multiple epigenetic changes that can have a profound effect on carcinogenesis have been identified. These epigenetic events, such as the methylation of promoter sequences in genes, are under active investigation. In this review we will describe a methylation event that occurs during the propagation of human mammary epithelial cells (HMEC) in culture and detail the accompanying genetic alterations that have been observed.