Granulocytic differentiation of HL-60 cells is not regulated by DNA de novo methylation

Traction stress analysis and modeling reveal that amoeboid migration in confined spaces is accompanied by expansive forces and requires the structural integrity of the membrane-cortex interactions

Leukocytes and tumor cells migrate via rapid shape changes in an amoeboid-like manner, distinct from mesenchymal cells such as fibroblasts. However, the mechanisms of how rapid shape changes are caused and how they lead to migration in the amoeboid mode are still unclear. In this study, we confined differentiated human promyelocytic leukemia cells between opposing surfaces of two pieces of polyacrylamide gels and characterized the mechanics of fibronectin-dependent mesenchymal versus fibronectin-independent amoeboid migration. On fibronectin-coated gels, the cells form lamellipodia and migrate mesenchymally. Whereas in the absence of cell-substrate adhesions through fibronectin, the same cells migrate by producing blebs and "chimneying" between the gel sheets. To identify the orientation and to quantify the magnitude of the traction forces, we found by traction force microscopy that expanding blebs push into the gels and generate anchoring stresses whose magnitude increases with decreasing gap size while the resulting migration speed is highest at an intermediate gap size. To understand why there exists such an optimal gap size for migration, we developed a computational model and showed that the chimneying speed depends on both the magnitude of intracellular pressure as well as the distribution of blebs around the cell periphery. The model also predicts that the optimal gap size increases with weakening cell membrane to actin cortex adhesion strength. We verified this prediction experimentally, by weakening the membrane-cortex adhesion strength using the ezrin inhibitor, baicalein. Thus, the chimneying mode of amoeboid migration requires a balance between intracellular pressure and membrane-cortex adhesion strength.

Adenosine analogs as possible differentiation-inducing agents against acute myeloid leukemia

Several adenosine analogs induce the functional and morphological differentiation of myelomonocytic leukemia cells. They can be classified into two types; i.e., those that do/do not require phosphorylation to induce the differentiation of leukemia cells. Neplanocin A, a potent S-adenosylhomocysteine hydrolase inhibitor, induces the differentiation of some leukemia cells without phosphorylation. On the other hand, deoxycoformycin (dCF), a potent adenosine deaminase inhibitor, also induces the myelomonocytic differentiation of leukemia cells when it is treated with deoxyadenosine (dAdo). This differentiation is inhibited by 5'-amino-deoxyadenosine, an inhibitor of (deoxy)adenosine kinase, suggesting that kinase-dependent phosphorylation is involved in the differentiation-inducing effect of dCF plus dAdo. Retinoids induce the differentiation of NB4 cells, a cell line derived from human promyelocytic leukemia. When used in combination with all-trans retinoic acid (ATRA), both NPA and dCF plus dAdo greatly enhance the granulocytic differentiation of NB4 cells. This enhancing effect is greatest when the cells are pretreated with NPA and then with ATRA. On the other hand, pre-exposure of NB4 cells to ATRA greatly potentiates the differentiation induced by dCF plus dAdo, while pretreatment with dCF plus dAdo before exposure to ATRA is less effective. The ATRA-induced differentiation of NB4 cells is effectively augmented by clinically applicable concentrations of these analogs. A clinical strategy that combines intermittent treatment with these analogs and a low dose of ATRA may increase the clinical response and decrease the adverse effects of ATRA.

Specificity of the G1 block induced by ethionine in human lymphocytes in vitro: a flow cytometric analysis

Ethionine is the ethyl analogue of the amino acid methionine. The agent is well known to have a weak demethylating activity. In addition, its capacity to reversibly block the cell cycle progression in G1 human lymphocytes (HL) without interfering with blastic transformation has been reported. In order to better understand the mechanism by which the agent is able to induce cell cycle block, experiments have been performed by using flow cytometry, in HL. In particular the hypothesis of the involvement of a specific target at the G0/G1 boundary was tested by treating HL at different times after blastic transformation. Starting from the 40th hour after PHA stimulation, ethionine loses its blocking capacity in such a way that cells challenged by the agent do not differ from controls in any one of the tested cell cycle-related parameters. We suggest the agent exerts its blocking activity at a specific stage of the transformation pathway.

Is there an association between an increase in c-myc RNA steady state levels and c-myc methylation in HL-60 cells treated with 3-deaza-(+/-)-aristeromycin, an indirect inhibitor of methylation?

Alteration in gene expression of the proto-oncogene c-myc in HL-60 cells is associated with differentiation of these cells. We have studied the steady state levels of c-myc transcripts, the levels of transmethylation metabolites S-adenosylmethionine and S-adenosyl-homocysteine and the methylation pattern of the c-myc gene after treatment of HL-60 cells with the transmethylation inhibitor and granulocytic inducer, 3-deaza-(+/-)-aristeromycin. A transient increase in c-myc RNA levels after 45 min of drug exposure was observed which was accompanied by changes in the ratio of transmethylation metabolites in both whole cells and nuclei. The changes in transmethylation metabolites in whole cells, although compatible with levels frequently associated with hypomethylation of cellular components, caused no changes in methylation of c-myc DNA sequences of the HL-60 cells as detected by HpaII or MspI digestion and Southern blotting.