p27Kip1 accumulation is associated with retinoic-induced neuroblastoma differentiation:: evidence of a decreased proteasome-dependent degradation

Development of human neuroblastoma is due to an arrest in the differentiation program of neural crest sympathoadrenal progenitor cells. However, neuroblastomas, as well as their derived cell lines, maintain the potentiality of terminal differentiation. We investigated the molecular mechanisms by which retinoic acid, a molecule introduced in clinical trials for chemotherapy, induces differentiation in neuroblastoma cell lines. Our findings demonstrate that the retinoic acid-dependent growth arrest of LAN-5 neuroblastoma cell line is associated to a very large accumulation (>tenfold) of p27(Kip1) protein, a cyclin-dependent kinase inhibitor; the protein binds and inhibits cyclin-dependent kinase 2, 4 and 6 activities, thus hampering pRb and p107 phosphorylation, p27(Kip1) build-up was observable as an early phenomenon (12-24 h) after retinoic exposure and resulted in a time-dependent accumulation of high quantities of a free p27(Kip1) form, Furthermore, retinoic treatment causes an increase of cyclin-dependent kinase 5 level and activity; however, immunoprecipitation studies proved the absence of interaction with p27(kip1). No noticeable variation of other components of G1 phase cell cycle engine ws observed. Pulse-chase experiments showed a remarkable elongation of p27(Kip1) half-life in retinoic-treated LAN-5, while no enhancement of p27(Kip1) gene expression and of the translational efficiency of its messenger RNA were demonstrated. In vivo degradation of p27(Kip1) was sensitive to two highly specific proteasome inhibitors, LLnL and lactacystin, while the calpain inhibitor II ALLM and the cysteine protease inhibitor E64 did not modify the level of the protein. LLnL treatment caused a very rapid (2 h) build-up of the Cdk inhibitor content and the accumulation of higher molecular weight anti-p27(Kip1) immunoreactive bands, which probably represent ubiquitinated forms of the protein. Finally, in vitro experiments demonstrated that extracts prepared from retinoic-treated LAN-5 cells degraded recombinant p27(Kip1) at a rate remarkably slower than the untreated cells. Our results indicate that retinoic acid strongly increases p27(Kip1) le, els by down-regulating the ubiquitin-proteasome p27(Kip1) degrading pathway.