Functional evidence for tumor-suppressor activity on chromosome 15 in human skin carcinoma cells and thrombospondin-1 as the potential suppressor
✍ Scribed by Petra Boukamp; Kerstin Bleuel; Susanne Popp; Vera Vormwald-Dogan; Norbert E. Fusenig
- Publisher
- John Wiley and Sons
- Year
- 1997
- Tongue
- English
- Weight
- 245 KB
- Volume
- 173
- Category
- Article
- ISSN
- 0021-9541
No coin nor oath required. For personal study only.
✦ Synopsis
During the development of skin cancer, normal kera-nign tumors. With later-passage HaCaT cells, which had accumulated a high number of chromosomal aber-tinocytes accumulate a number of genetic aberrations that finally allow a cell to clonally expand and form a rations, the same oncogene caused malignant conversion in all clones tested (Boukamp et al., 1995). Thus, tumor. At present, the best documented genetic changes relevant for nonmelanoma skin carcinomas are muta-the later-passage HaCaT cells had gained one or more genetic defects that were able to cooperate with the tional inactivation of the p53 tumor-suppressor gene (for review, see Ziegler et al., 1994), loss of material from oncogenic potential of the ras gene, and only this combination allowed the cells to become malignant tumori-chromosome 9 (Quinn et al., 1994), and ras oncogene activation (for review, see Ananthaswamy and Pierceall, genic.
In search of these genetic defects, we found that chro-1992).
mosome 15 was underrepresented in all malignant IN VITRO MODEL OF SKIN ''late-passage'' clones as compared with the non-or be-CARCINOGENESIS nign-tumorigenic ''early-passage'' clones (Boukamp et al., 1995). Similarly, in the stable nontumorigenic Ha-To determine functional consequences of these ge-CaT cells, the number of chromosome 15 remained connetic changes, we have developed an in vitro model stantly high, with 4 -5 copies up to more than 300 pasof human skin carcinogenesis, i.e., the spontaneously sages (Boukamp et al.,1997). immortalized HaCaT cells (Boukamp et al., 1988). With these cells, we were able to substantiate the role of p53 ALTERATIONS IN CHROMOSOME 15 mutation as a primary event in skin carcinogenesis IN SKIN CARCINOMA CELLS (Lehman et al., 1993). In HaCaT cells, both alleles were This finding prompted us to investigate the status of mutated in CC sites from the early beginning, and, chromosome 15 in cell lines established from human typically, one allele carried a CC-to-TT double-base squamous cell carcinoma (SCCs). In 4 of 5 lines, chrochange indicative for UV-B-induced mutations. In addimosome 15 was not present in two normal copies. In 3 tion, HaCaT cells showed loss of one copy of chromolines, one copy was involved in a translocation; from somes 9p, 3p, and 4p at this early time point. By reinstudies with yac clones spanning the subcentromeric troducing extra copies of these chromosomes via microregion of chromosome 15, part of this chromosome may cell-mediated chromosome transfer (MMCT; Saxon et have been lost (Popp and Boukamp, unpublished real., 1985), we were able to show that chromosome 9 sults). In the fourth line, one complete copy was missand chromosome 4 had no effect on the growth of Haing, and loss was seen as early as in explant culture CaT cells. However, introduction of chromosome 3 re- (Boukamp et al., 1982). Thus, at least in this line, loss sulted in senescence of the immortal HaCaT cells (Bouof chromosome 15 was not a tissue culture artifact but kamp, unpublished results).
was acquired during tumor development in the patient. Surprisingly, the ras oncogene also had no dominant effect in this model system. Its oncogenic potential de-FUNCTIONAL CONSEQUENCE OF LOSS pended on the genetic background of the HaCaT cells,
OF CHROMOSOME 15 which was demonstrated in a set of experiments in
To investigate the role of a loss of chromosome 15 in which we introduced the ras oncogene not via transfecthe SCL-I cells, we introduced an extra copy of chromotion but via MMCT. This approach was advantageous some 15 via microcell-mediated chromosome transfer. in two ways: we avoided (a) different amounts of inte-The resulting microcell chr15 hybrid cells were analyzed grated ras oncogene and (b) different integration sites for in vitro and in vivo growth. In culture, the populaof the gene in the individual recipient cells. When earlytion doubling time remained similar to that of the papassage HaCaT cells (with only few genetic changes) were used as recipient cells, the ras oncogene did not lead to tumorigenicity at all or, with two exceptions, *Correspondence to: Petra Boukamp, Department of Carcinogen- the cells developed small tumors that, as shown by esis and