Axel H. Schönthal

Department of Molecular Microbiology and Immunology, K. Norris Jr. Comprehensive Cancer Center, University of Southern California, 2011 Zonal Ave., HMR-405, Los Angeles, CA 90033

Received 8/25/98 Accepted 12/2/98

5. PP2A AS A TUMOR SUPPRESSOR

The finding that okadaic acid is a tumor promoter, combined with the observation that this compound efficiently inhibits PP2A, has led to the suggestion that PP2A, and potentially some other phosphatases, may function as tumor suppressors (60). It is thought that the tumor suppressing function could be accomplished by the enhanced dephosphorylation of activated kinase cascades, which would revert oncoprotein-activated signaling pathways back to their inactive state. As many oncogene products cause the sustained activation of growth-regulatory kinase cascades, it is a distinct possibility that increased serine/threonine protein phosphatase activity might counteract elevated levels of protein phosphorylation and block cellular transformation (48).

Further support for such a negative role of phosphatases in growth-regulatory signal transduction pathways came from observations that treatment of cells with okadaic acid led to the increased expression of several proto-oncogenes (see detailed references in (12)). Elevated expression of such genes had been shown before to contribute to cellular transformation (1). Thus, these results suggested that the respective okadaic acid sensitive phosphatases contribute to the repression of these proto-oncogenes in normal cells.

Moreover, when added to cells synchronized in the S phase of the cell cycle, okadaic acid caused an increase in the enzymatic activity of cyclin-dependent kinase (histone H1 kinase), an enzyme that is necessary for cell cycle progression. The drug also stimulated premature mitosis and increased the phosphorylation of mitosis-specific proteins (see detailed references in (12). When added at low concentrations, okadaic acid caused quiescent fibroblasts to progress to S phase of the cell cycle (98); and in thyroid cells, the drug increased the fraction of thyrotropin-stimulated quiescent cells entering S phase (99). Further, a link between protein phosphatases and cellular transformation has been established by the finding that small DNA tumor viruses, such as Simian virus 40 (SV40) and Polyoma virus, synthesize proteins (small and medium T antigens) that bind to and inhibit PP2A (100). The currently available evidence indicates that alteration of phosphatase activity and subsequent changes in phosphorylation levels is a crucial step in transformation by these viruses (55, 100). Another viral protein, E4orf4 of Adenovirus, has been found to associate with PP2A as well (101).

Further evidence for the involvement of phosphatases in the neoplastic process has been provided by the finding that certain cis-platin resistant human cancer cells are resistant to growth inhibition by okadaic acid, and exhibit increased phosphorylation of certain nuclear proteins (102, 103). Resistance to okadaic acid has also been observed in cells that exhibit the multidrug resistance (mdr) phenotype (104-108). Mdr cells were established by chronic exposure of cells to increasing concentrations of okadaic acid. In this case, two different mechanisms were found to generate the mdr phenotype.

First, mdr cells had amplified the gene encoding the 170-kDa P-glycoprotein, which is a drug efflux pump with broad specificity, i.e. it is capable of extruding intracellular anticancer agents of diverse structures and mechanisms of action (105, 106). Consequently, these cells were not only resistant to okadaic acid and related phosphatase-inhibitory compounds, but also to other structurally unrelated anticancer drugs, such as vinblastine, taxol, or cisplatin. In addition, the activity of P-glycoprotein appears to be regulated via its phosphorylation status, which is increased in response to treatment of cells with okadaic (109). It should be noted, however, that the increased activity of the P-glycoprotein pump could not be demonstrated in all okadaic acid resistant cell types; in some cells no differences could be found in the accumulation or efflux of okadaic acid between drug resistant and normal cells (104, 107, 110).

A second mechanism that was found to generate okadaic acid resistant cell lines is the mutational alteration of the PP2A catalytic subunit. Upon sequencing of the PP2A gene from okadaic acid resistant hamster cells, a point mutation was found that resulted in the exchange of cysteine 269 for glycine (111). This mutation resulted in a PP2A protein that was much more resistant to inhibition by okadaic acid than the wild type protein. Further mutational analysis of this region established that the amino acids 265 to 269 are critical for inhibition of PP2A by okadaic acid (112). Mdr cell lines that harbored such a mutation of PP2A were found to express the mdr phenotype in a stable fashion. In contrast, cell lines without PP2A mutations but with amplified P-glycoprotein, tended to loose the mdr phenotype after cessation of long-term drug exposure (110).

In addition to gene amplification, the treatment of cells with okadaic acid has also generated other genetic changes in cultured cells, such as mutations endowing diphtheria-toxin resistance, sister chromatid exchange in the presence of bromodeoxyuridine, loss of exogenous transforming oncogenes, and minisatellite mutations (26, 113-116). Although the molecular mechanisms underlying this genotoxic activity of okadaic acid have not been elucidated, it is suspected that the alteration of the phosphorylation status of cellular proteins, and the resulting changes in the gene expression pattern, might be a crucial epigenetic event contributing to these processes. In this regard, it is important to note that okadaic acid induces elevated and sustained expression of the c-fos proto-oncogene (63, 105). Since c-fos has been shown to increase the spontaneous level of chromosomal aberrations (117-119), it is conceivable that okadaic acid may stimulate these processes through its continuous activation of c-fos expression. Furthermore, since okadaic acid exerts its effect on c-fos expression through inhibition of PP2A (120, 121), one could envision that PP2A, through its negative effects on the c-fos gene, may contribute to the maintenance of genomic integrity. Thus, these observations give further credence to the idea that PP2A indeed may act as a tumor suppressor.