Agents that down-regulate or inhibit protein kinase C circumvent resistance to 1-beta-D-arabinofuranosylcytosine-induced apoptosis in human leukemia cells that overexpress Bcl-2.

The effects of the non-tumor-promoting protein kinase C (PKC) activator bryostatin 1 and the PKC inhibitors staurosporine and UCN-01 were examined with respect to modulation of 1-[beta-D-arabinofuranosyl]cytosine (ara-C)-induced apoptosis in human myeloid leukemia cells (HL-60) overexpressing the antiapoptotic protein Bcl-2. HL-60/Bcl-2 cells displayed a 5-fold increase in Bcl-2 protein compared with empty-vector counter-parts (HL-60/pCEP4) but comparable levels of Bax, Mcl-1, and Bcl-xL. After exposure to an equimolar concentration of ara-C (10 microM for 6 hr), HL-60/Bcl-2 cells were significantly less susceptible to apoptosis, DNA fragmentation, and loss of clonogenicity than HL-60/pCEP4 cells. The protective effect of increased Bcl-2 expression was manifested by a failure of ara-C to induce activation/cleavage of the Yama protease (CPP32; caspase-3) and degradation of one of its substrates, poly(ADP-ribose)polymerase to an 85-kDa cleavage product. When HL-60/Bcl-2 cells were preincubated with bryostatin 1 (10 nM; 24 hr) or coincubated with either staurosporine (50 nM; 6 hr) or UCN-01 (300 nM; 6 hr) after a 1-hr preincubation, exposures that exerted minimal effects alone, ara-C-induced apoptosis and DNA fragmentation were restored to levels equivalent to, or greater than, those observed in empty-vector controls. These events were accompanied by restoration of the ability of ara-C to induce CPP32 cleavage and activation, poly(ADP-ribose) polymerase degradation, and inhibition of colony formation. Western analysis of Bcl-2 protein obtained from overexpressing cells treated with bryostatin 1, staurosporine, or UCN-01 revealed the appearance of a slowly migrating species and a general broadening of the protein band, effects that were insensitive to the protein synthesis inhibitor cycloheximide. Alterations in Bcl-2 protein mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis were reversed by treatment of lysates with alkaline phosphatase or protein phosphatase 2A; actions of the latter were blocked by the specific phosphatase inhibitor okadaic acid. In vivo labeling studies of Bcl-2 protein demonstrated increased incorporation of [32PO4]orthophosphate in drug-treated cells. Last, phosphorylated Bcl-2 failed to display decreased binding to the proapoptotic protein Bax. Collectively, these findings indicate that bryostatin 1, which down-regulates PKC, and staurosporine and UCN-01, which directly inhibit the enzyme, circumvent resistance of Bcl-2-overexpressing leukemic cells to ara-C-induced apoptosis and activation of the protease cascade. They also raise the possibility that modulation of Bcl-2 phosphorylation status contributes to this effect.

Modulation of the expression of Bcl-2 and related proteins in human leukemia cells by protein kinase C activators: relationship to effects on 1-[beta-D-arabinofuranosyl]cytosine-induced apoptosis.

We have previously reported that pretreatment of HL-60 human promyelocytic leukemia cells with the non-tumor-promoting protein kinase C (PKC) activator bryostatin 1 potentiates induction of apoptosis by the antimetabolite 1-[beta-D-arabinofuranosyl]cytosine (ara-C) (Biochem Pharmacol 47:839,1994). To determine whether this phenomenon results from altered expression of Bcl-2 or related proteins, Northern and Western analysis was employed to assess the effects of bryostatin 1 and other PKC activators on steady-state levels of Bcl-2, Bax, Bcl-x, and Mcl-1 mRNA and protein. Pretreatment of cells for 24 h with 10 nM bryostatin 1, or, to a lesser extent, the stage-1 tumor-promoter phorbol dibutyrate (PDB) significantly potentiated apoptosis induced by ara-C (100 microM; 6 h); in contrast, equivalent exposure to the stage-2 tumor promoter, mezerein (MZN), which, unlike bryostatin 1, is a potent inducer of differentiation in this cell line, failed to modify ara-C-related cell death. Neither bryostatin 1 nor PDB altered expression of bcl-2/Bcl-2 over this time frame. In contrast, MZN down-regulated bcl-2 mRNA levels, but this effect was not accompanied by altered expression of Bcl-2 protein. None of the PKC activators modified expression of Bax or Bcl-x(L) mRNA or protein; levels of Bcl-x(S) were undetectable in both treated and untreated cells. However, expression of Mcl-1 mRNA and protein increased modestly after treatment with either bryostatin 1 or PDB, and to a greater extent following exposure to MZN. Combined treatment of cells with bryostatin 1 and MZN resulted in undiminished potentiation of ara-C-mediated apoptosis and by antagonism of cellular maturation. These effects were accompanied by unaltered expression of Bcl-2, Bax, and Bcl-x(L), and by a further increase in Mcl-1 protein levels. When cells were co-incubated with bryostatin 1 and calcium ionophore (A23187), an identical pattern of expression of Bcl-2 family members was observed, despite the loss of bryostatin 1's capacity to potentiate apoptosis, and the restoration of its ability to induce differentiation. Finally, treatment of cells with bryostatin 1+/-ara-C (but not ara-C alone) resulted in a diffuse broadening of the Bcl-2 protein band, whereas exposure of cells to taxol (250 nM, 6 h) led to the appearance of a distinct Bcl-2 species with reduced mobility, phenomena compatible with protein phosphorylation. Together, these findings indicate that the ability of bryostatin 1 to facilitate drug-induced apoptosis in human myeloid leukemia cells involves factors other than quantitative changes in the expression of Bcl-2 family members, and raise the possibility that qualitative alterations in the Bcl-2 protein, such as phosphorylation status, may contribute to this capacity. They also suggest that increased expression of Mcl-1 occurs early in the pre-commitment stage of myeloid cell differentiation, and that this event does not protect cells from drug-induced apoptosis.

Modulation of 1-[beta-D-arabinofuranosyl] cytosine-induced apoptosis in human myeloid leukemia cells by staurosporine and other pharmacological inhibitors of protein kinase C.

We have examined the effects of both nonspecific and highly selective pharmacological inhibitors of protein kinase C (PKC) on the capacity of a 6-h exposure to 1-[beta-D-arabinofuranosyl]cytosine (ara-C; 10 microM) to induce apoptotic DNA fragmentation and cell death in the human myeloid leukemia cell lines HL-60 and U937. Staurosporine, a highly potent, nonspecific inhibitor of PKC (20-50 nM), uniquely potentiated ara-C-related degradation of DNA to oligonucleosomal fragments in both cell lines (i.e., 2- to 3-fold), but was ineffective when given alone at these concentrations. In contrast, co-administration of the nonspecific PKC inhibitor H7 and two highly selective PKC inhibitors, calphostin C and chelerythrine, also increased the extent of DNA fragmentation observed in ara-C-treated cells, but these effects were evident only at inhibitor concentrations that were by themselves sufficient to induce DNA damage. Agarose gel electrophoresis demonstrated that cells co-exposed to staurosporine and ara-C exhibited considerably more pronounced internucleosomal DNA cleavage than did cells exposed to ara-C alone; moreover, this effect was suppressed by Zn2+ (1 mM) and the permeant Ca2+ chelator BAPTA-AM (50 microM). Potentiation of ara-C-related DNA fragmentation by subeffective concentrations of staurosporine was accompanied by a pronounced increase in the morphological features characteristic of apoptosis. A synergistic interaction between staurosporine and ara-C with respect to inhibition of clonogenicity in both HL-60 and U937 cells was demonstrated by median dose-effect analysis. The actions of staurosporine did not result from enhanced ara-C metabolism, as preincubation of cells with concentrations of this agent that potentiated ara-C actions (e.g., 20-50 nM) did not increase intracellular levels of the lethal metabolite ara-CTP. Lastly, preexposure of HL-60 and U937 cells to staurosporine did not block ara-C-mediated upregulation of c-jun, an oncogene whose increased expression has been temporally associated with ara-C-induced apoptosis. Together, these findings indicate that staurosporine exhibits a unique pattern of potentiation of ara-C-related apoptosis in human myeloid leukemias, and provide a rationale for exploring the antileukemic potential of this combination regimen.