In tight collaboration with the protein laboratory of the FLI we used 1- and 2-D protein electrophoresis, N-terminal protein sequencing and protein or peptide mass spectrometry (MALDI-TOF, ESI-MS) to identify proteins that interact with replication factors such as Cdc45p and TopBP1. We are also interested in proteins that coimmunoprecipitate with the tumor suppressor protein p53. With the latter approach we identified some 200 potential protein partners of p53 that include several replication factors. Even if only 10 percent of the found proteins represent true interactors, this would indicate an impressing p53-based regulatory network. Out of these interactors the chaperone and peptidyl-prolyl isomerase cyclophilin 18 (Cyp18) has been studied more extensively, mainly because it is the main target of the immunosuppressant and frequently prescribed drug cyclosporine A (CsA).
About 10 percent of the CsA-treated patients develop skin cancers for hitherto unknown reasons. Cyp18 binds to proline 72 of p53 (see Figure) and decreases its transcriptional activity. In line with this, the antagonist of cyp18, CsA, induces a cell cycle arrest particularly in cells that carry the proline 72 allele of p53, but much less so in those carrying the polymorphic variant arginine 72. CsA also increases apoptosis in p53-containing cells but not in cells lacking this tumor suppressor. Taken together, CsA stimulates rather than inhibits p53's pro-apoptotic functions.
How then can CsA stimulate skin cancer formation in immunosuppressed patients after organ transplantation? One possibility is that the immunosuppressed state alone is responsible for the increased cancer incidence. This, however, is not very likely since immuno-compromised nude mice develop similar symptoms after CsA application and patients treated with another immunosuppressive drug, rapamycin, do not develop skin cancer. Hence it is likely, that long-term treatment with CsA may induce an upregulation of Cyp18, which in accordance with our results, decreases the amount of active p53 particularly when the proline 72 variant is expressed. We are currently trying to evaluate this working hypothesis.
Model of the p53-Cyp18 interaction: Biochemical data and structural considerations support a direct p53-Cyp18 interaction that requires the catalytic site of Cyp18 and the proline-rich area centered at Pro71-Pro72 of human p53. On the left hand side there is an overall model of p53 binding to Cyp18, on the right hand side we depict a close-up view on Cyp18's binding to the proline-rich region of p53. We gratefully acknowledge Oliver Ohlenschläger (NMR group of theFLI) and Gunter Fischer as well as Cordelia Schiene-Fischer (MPG for Protein Folding, Halle) for their contributions to the establishment of this model.
|
