Barbara A. Hocevar

Carcinogenesis is a multistep process involving mutation and the subsequent selective clonal expansion of the mutated cell. Chemical and physical agents including those that induce reative oxygen species can induce and/or modulate this multistep process. Several modes of action by which carcinogens induce cancer have been identified, including through production of reactive oxygen species (ROS). Oxidative damage to cellular macromolecules can arise through overproduction of ROS and faulty antioxidant and/or DNA repair mechanisms. In addition, ROS can stimulate signal transduction pathways and lead to activation of key transcription factors such as Nrf2 and NF-kappaB. The resultant altered gene expression patterns evoked by ROS contribute to the carcinogenesis process. Recent evidence demonstrates an association between a number of single nucleotide polymorphisms (SNPs) in oxidative DNA repair genes and antioxidant genes with human cancer susceptibility. These aspects of ROS biology will be discussed in the context of their relationship to carcinogenesis.

The promyelocytic leukemia (HL60) cell line differentiates into monocyte-like cells after treatment with phorbol dibutyrate (PBt2). In contrast, bryostatin 1 (bryo), a structurally distinct protein kinase C (PKC) activator, does not induce differentiation and blocks the cytostatic effect of PBt2. The divergent responses to these agents correlate with activation of a PKC-like activity at the nucleus in response to bryo but not PBt2 (Fields, A. P., Pettit, G. R., and May, W.S. (1988) J. Biol. Chem. 263, 8253-8260). In the present study, this nuclear PKC-like activity (termed PKCn) was isolated from HL60 cells and shown to phosphorylate its known nuclear substrate, lamin B. PKCn-mediated phosphorylation of nuclear envelope-associated lamin B in vitro is calcium-dependent and is stimulated by bryo and 1,2-dioctanoylglycerol (DiC8), but not PBt2. In contrast, PKCn-mediated phosphorylation of histone IIIS is stimulated equally by all three activators. PKCn mediates calcium- and phosphatidylserine-dependent phosphorylation of both histone IIIS and partially purified lamin B. PKCn activity can be inhibited by an anti-PKC monoclonal antibody which specifically inhibits PKC. Isotype-specific PKC antibodies identify PKCn as beta II-PKC. Immunoblot analysis indicates that HL60 cells express both alpha- and beta II-PKC but no beta I- or gamma-PKC. Treatment of intact cells with bryo for 30 min leads to complete translocation of both alpha- and beta II-PKC from the cytosol to the membrane fractions. Approximately 8-10% of the total beta II-PKC (and less than 0.3% of the alpha-PKC) is found associated with the nuclear membrane of bryo-treated cells. In contrast, PBt2 treatment leads to complete translocation of alpha-PKC, but only partial translocation of beta II-PKC to the plasma membrane fraction. Neither PKC isotype is found associated with the nuclear membrane of PBt2-treated cells. These data demonstrate that alpha- and beta II-PKC differ with respect to activator responsiveness, intracellular distribution, and substrate specificity and indicate that their selective activation at distinct intracellular sites, including the nucleus, can have a dramatic effect on resulting cellular responses.

Multisite phosphorylation of the nuclear lamins is thought to regulate the process of mitotic nuclear envelope breakdown in vivo. Here we investigate the involvement of two proposed human mitotic lamin kinases, beta II protein kinase C (PKC) and p34cdc2/cyclin B kinase, in human lamin B1 phosphorylation in vitro and in intact cells. We find that both kinases can phosphorylate purified soluble lamin B at similar rates. However, beta II PKC phosphorylates interphase nuclear envelope lamin B at more than 200 times the rate of human p34cdc2/cyclin B kinase. beta II PKC-mediated phosphorylation of lamin B is confined to two sites, Ser395 and Ser405, within the carboxyl-terminal domain, whereas human p34cdc2/cyclin B kinase phosphorylates a single site, Ser23, in the amino-terminal domain. A second potential p34cdc2/cyclin B kinase site within the carboxyl-terminal domain, Ser393, is not phosphorylated by human p34cdc2/cyclin B kinase. However, invertebrate p34cdc2/cyclin B kinase from sea star exhibits a different specificity, phosphorylating both amino- and carboxyl-terminal sites. Mitotic human lamin B from intact cells is phosphorylated predominantly in its carboxyl-terminal domain. Comparative tryptic phosphopeptide mapping demonstrates that the beta II PKC site, Ser405, is a prominent target of mitotic lamin B phosphorylation in vivo. beta II PKC translocates to the nucleus during the G2/M phase of cell cycle concomitant with phosphorylation of Ser405, indicating a physiologic role for nuclear beta II PKC activation in mitotic lamin B phosphorylation in vivo. The presence of phosphorylation sites within the carboxyl-terminal domain of mitotic lamin B which are not phosphorylated by either beta II PKC or p34cdc2/cyclin B kinase suggests the involvement of other lamin kinase(s) in G2/M phase lamin B phosphorylation.