The multi-step process of human skin carcinogenesis: A role for p53, cyclin D1, hTERT, p16, and TSP-1

• Published in: European journal of cell biology 2007-12, Vol.86 (11), p.763-780
• Main Authors: Burnworth, Bettina, Arendt, Susanne, Muffler, Sonja, Steinkraus, Volker, Bröcker, Eva B., Birek, Catalina, Hartschuh, Wolfgang, Jauch, Anna, Boukamp, Petra
• Format: Article
• Language: English
• Subjects: Animals; Carcinoma, Squamous Cell - enzymology; Cell Cycle; Chromosomes, Human, Pair 15; Cyclin D1; Cyclin D1 - metabolism; Cyclin-Dependent Kinase Inhibitor p16 - metabolism; Down-Regulation; Humans; Keratoacanthoma; Keratoacanthoma - enzymology; Mice; Mice, Nude; Models, Biological; Mutation - genetics; P16; P53; Precancerous Conditions - pathology; Skin Neoplasms - pathology; Squamous cell carcinoma; Telomerase - metabolism; Thrombospondin; Thrombospondin 1 - deficiency; Thrombospondin 1 - metabolism; Tumor Suppressor Protein p53 - metabolism
• ISSN: 0171-9335; 1618-1298
• DOI: 10.1016/j.ejcb.2006.11.002

As proposed by Hanahan and Weinberg (2000. Cell 100, 57–70) carcinogenesis requires crucial events such as (i) genomic instability, (ii) cell cycle deregulation, (iii) induction of a telomere length maintenance mechanism, and (iv) an angiogenic switch. By comparing the expression of p53, cyclin D1, p16, hTERT, and TSP-1 in spontaneously regressing keratoacanthoma (KA) as a paradigm of early neoplasia, with malignant invasive cutaneous squamous cell carcinoma (SCC) as a paradigm of advanced tumour development, we are now able to assign the changes in the expression of these proteins to specific stages and allocate them to defined roles in the multi-step process of skin carcinogenesis. We show that mutational inactivation of the p53 gene, and with that the onset of genomic instability is the earliest event. Individual p53-positive cells are already seen in “normal” skin, and 3/5 actinic keratoses (AKs), 5/22 KAs, and 13/23 SCCs contain p53-positive patches. Cell cycle deregulation was indicated by the overexpression of the cell cycle regulator cyclin D1, as well as by the loss of the cell cycle inhibitor p16. Interestingly, overexpression of cyclin D1 – observed in 80% of KAs and SCCs, respectively – showed a cell cycle-independent function in HaCaT cell transplants on nude mice. Cyclin D1 overexpression was associated with a massive inflammatory response, finally leading to tissue destruction. Loss of the cell cycle inhibitor p16, on the other hand, correlated with SCCs. Thus, it is tempting to suggest that overexpression of cyclin D1 is an early change that in addition to growth stimulation leads to an altered epithelial–mesenchymal interaction, while functional p16 is able to control this deregulated growth and needs to be eliminated for malignant progression. Another requirement for uncontrolled growth is the inhibition of telomere erosion by up-regulating telomerase activity. As measured by hTERT protein expression, all of the KAs and SCCs studied were positive, with a similar distribution of the protein in both groups and an expression pattern resembling that of normal epidermis. Thus, telomerase may not need to be increased significantly in skin carcinomas. Finally, we show that the angiogenesis inhibitor TSP-1 is strongly expressed in most KAs, and mainly by the tumour cells, while in SCCs the generally weak expression is restricted to the tumour–stroma. Furthermore, we provide evidence that the loss of a copy of chromosome 15 is responsible for redu