Embryonic stem cells as a model for studying regulation of cellular differentiation

Mouse embryonic stem (ES) cells can be differentiated in vitro into near homogeneous populations of both neurons and skeletal muscle as well as other cell types. We previously showed that treatment of pluripotent ES cells with retinoic acid (RA) induced differentiation into highly enriched populations of gamma-aminobutyric acid (GABA) expressing neurons. The reasons for generation of only GABA neurons as opposed to other neuronal cell types were not known. We have extended our previous work and now show that with RA induction of ES cells we not only obtain GABA neurons, but also dopaminergic neurons. Critical for the production of dopaminergic neurons after RA induction was the post-induction plating conditions used. No dopaminergic neurons were detected if cells were plated in serum-free media optimized for neuronal survival. However, significant numbers of dopamine neurons could be detected when cells were plated in media containing fetal calf serum. These observations support the conclusion that RA acts as a general neural inducing agent and that conditions post-induction either selectively support survival of a particular class of neuronal cells or that the conditions post-induction actually further instruct cells to differentiate into different types of neurons.     

The human promyelocytic HL60 cell line: a model of myeloid cell differentiation using dimethylsulphoxide, phorbol ester and butyrate.

The release of the reactive oxygen species that accompanies the oxidative burst was studied in HL60 cells differentiated with either dimethylsulphoxide, butyrate or phorbol myristate acetate in order to establish the extent to which differentiated cells are phenotypically similar to human neutrophils, monocytes and macrophages. When phorbol myristate acetate was used as a stimulus, the rates of superoxide production by dimethylsulphoxide and butyrate differentiated HL60 cells was not significantly different from those observed in neutrophils and monocytes isolated from normal peripheral blood. Similar results were obtained when luminol-dependent chemiluminescence was measured in the presence of horseradish peroxidase using phorbol myristate acetate as the stimulus. However, in the absence of horseradish peroxidase, the luminol-dependent chemiluminescence in the dimethylsulphoxide and butyrate-differentiated HL60 cells was significantly lower than that of the control cells isolated from human blood, reflecting the absence of myeloperoxidase in the differentiated cells. In contrast, HL60 cells differentiated by phorbol myristate acetate failed to show any increased generation of superoxide or luminol-dependent chemiluminescence upon stimulation. Impaired release of lysosomal enzymes by the chemically differentiated cells suggests impairments in the extent of differentiation resulting in cells with defective azurophilic degranulation processes. It is concluded that HL60 cells differentiated by the above agents are somewhat controversial models of promyelocyte differentiation into typical neutrophilic, monocytic and macrophage-like cells.     

Dephosphorylation of the retinoblastoma protein during differentiation of HL60 cells.

Immunoprecipitated retinoblastoma protein from HL60 cells migrated as a series of bands during electrophoresis. The heterogeneity appeared to be generated by phosphorylation of the retinoblastoma protein. Treatment of the cells with the phorbol ester, tetradecanoyl phorbol acetate (TPA), resulted in both a loss of the heterogeneity of the pRB species and a significant decrease in the level of pRB phosphorylation. These changes accompanied differentiation of the HL60 cells into macrophages. Treatment of the cells with dibutyryl cAMP also resulted in dephosphorylation of pRB as well as cell cycle arrest, although no recognizable differentiation occurred. These results are consistent with a model in which TPA and dibutyryl cAMP dependent pathways can activate pRB by altering its phosphorylation.     

Human EVI9, a homologue of the mouse myeloid leukemia gene, is expressed in the hematopoietic progenitors and down-regulated during myeloid differentiation of HL60 cells

Evi9, a common site of retroviral integration in BXH2 murine myeloid leukemias, encodes a C2H2 zinc finger protein and is overexpressed in these leukemic cells. To investigate a possible role of EVI9 in the human hematopoietic system, we isolated the cDNA clone of the human homologue. Human EVI9, located on the chromosome 2p13 region, contains an open reading frame of 797 amino acids that is 98.7% identical to the mouse protein. RT-PCR analysis of purified human hematopoietic cells showed that EVI9 is expressed in CD34-positive myeloid precursors, B cells, monocytes, and megakaryocytes, but only weakly in T lymphocytes, suggesting that EVI9 may play an important role in hematopoiesis. Furthermore, EVI9 was down-regulated during myeloid differentiation of HL60 cells induced by all-trans-retinoic acid, whereas the expression remained during monocytic differentiation induced by phorbol 12-myristate 13-acetate. These results indicate a distinct role for EVI9 in human hematopoietic cells and suggest that EVI9 may cause leukemia through inhibition of myeloid differentiation.     

A system for monocytic differentiation of leukemic cells HL 60 by a short exposure to 1,25-dihydroxycholecalciferol

The human promyelocytic cell line HL 60 can be induced to differentiate toward more mature myeloid or monocytic forms by a variety of agents. This process is thought to require several days of exposure to the inducer, thus making it difficult to identify the early cellular changes which are fundamental to the differentiation program, and to relate the induction to phases of the cell cycle. In order to study the kinetics of leukemic cell differentiation we have developed a system for the induction of rapid monocytic maturation in a subpopulation of HL 60 cells. The cells are exposed to 10(-7) M 1,25-dihydroxycholecalciferol for 4 hr in serum-free medium. Subsequent incubation in a complete medium results in cellular differentiation recognizable by several criteria (phagocytosis, nonspecific esterase reaction, adherence to substratum, cell morphology) beginning at 10 hr from the exposure to the inducer. Approximately 20 hr later 30-40% of the cells in culture show the differentiated phenotype and are capable of phagocytosis. The proportion of differentiated cells in culture decreases thereafter. This system has been utilized to study the expression of c-myc oncogene in relation to the kinetics of maturation, and it was found that the inhibition of the expression of this gene precedes the onset of phenotypic differentiation by approximately 8 hr, is transient, and is accompanied by a brief retardation of cell proliferation, which resumes the normal rate within 24 hr of the exposure to the inducer.     

Oncostatin M is a differentiation factor for myeloid leukemia cells.

Oncostatin M (OSM) is a 28-kDa glycoprotein produced by stimulated macrophages and T lymphocytes that inhibits the proliferation of a number of different cell lines derived from solid tumors. Analysis of both amino acid sequence and gene structure has demonstrated that OSM is a member of a cytokine family that includes leukemia inhibitory factor (LIF), IL-6, and granulocyte colony-stimulating factor (G-CSF). We demonstrate that, like LIF, IL-6 and G-CSF, OSM can induce the differentiation of the myeloblastic M1 murine leukemia cells into macrophage-like cells. The morphologic and functional changes induced by OSM are more similar to those observed with LIF and IL-6 than those induced with G-CSF. OSM can also induce the differentiation of the histiocytic U937 human leukemia cells in the presence of granulocyte-macrophage CSF, a property shared with LIF and IL-6. In murine M1 cells, binding of labeled OSM is completely inhibited by excess LIF or OSM, reflecting the binding of OSM to the high affinity form of the murine LIF receptor. In contrast, the binding of labeled OSM to human U937 leukemia cells is inhibited by OSM, but the inhibition by LIF is significantly less. These results suggest that, in human leukemia cells, OSM may act through the LIF receptor and an OSM-specific receptor. The existence of an OSM-specific receptor was confirmed by both growth inhibition and competition binding assays on A375 human melanoma cells. The growth of human A375 cells was inhibited by OSM and IL-6 but not LIF or G-CSF. Neither LIF, G-CSF, nor IL-6 could compete with the binding of labeled OSM to A375 cells.     

Cdc6 is a rate-limiting factor for proliferative capacity during HL60 cell differentiation

The DNA replication (or origin) licensing pathway represents a critical step in cell proliferation control downstream of growth signalling pathways. Repression of origin licensing through down-regulation of the MCM licensing factors (Mcm2-7) is emerging as a ubiquitous route for lowering proliferative capacity as metazoan cells exit the cell division cycle into quiescent, terminally differentiated and senescent "out-of-cycle" states. Using the HL60 monocyte/macrophage differentiation model system and a cell-free DNA replication assay, we have undertaken direct biochemical investigations of the coupling of origin licensing to the differentiation process. Our data show that down-regulation of the MCM loading factor Cdc6 acts as a molecular switch that triggers loss of proliferative capacity during early engagement of the somatic differentiation programme. Consequently, addition of recombinant Cdc6 protein to in vitro replication reactions restores DNA replication competence in nuclei prepared from differentiating cells. Differentiating HL60 cells over-expressing either wild-type Cdc6 or a CDK phosphorylation-resistant Cdc6 mutant protein (Cdc6A4) exhibit an extended period of cell proliferation compared to mock-infected cells. Notably, differentiating HL60 cells over-expressing the Cdc6A4 mutant fail to down-regulate Cdc6 protein levels, suggesting that CDK phosphorylation of Cdc6 is linked to its down-regulation during differentiation and the concomitant decrease in cell proliferation. In this experimental model, Cdc6 therefore plays a key role in the sequential molecular events leading to repression of origin licensing and loss of proliferative capacity during execution of the differentiation programme.     

Differential effects of c-myb and c-fes antisense oligodeoxynucleotides on granulocytic differentiation of human myeloid leukemia HL60 cells

To gain some insight into the role of c-myb and c-fes in myeloid differentiation, the authors have analyzed the ability of HL60 cells to differentiate in response to several different inducers after inhibition of c-myb and c-fes function. This function has been inhibited almost completely by using deoxynucleotides complementary to two 18-nucleotide sequences of c-myb and c-fes encoding mRNA. After 5 days in culture, in several separate experiments with different oligomer preparations, more than 90% growth inhibition was observed in c-myb antisense-treated HL60 cells. At this time, independent of the differentiation inducer used, c-myb antisense-treated HL60 cells differentiate only along the monocytic pathway, whereas in sense oligomer-treated cultures, retinoic acid and dimethyl sulfoxide induced granulocytic differentiation. No perturbation of the HL60 cell growth was observed after 5 days of treatment with antisense c-fes oligomer. However, induction to granulocytic differentiation by retinoic acid and dimethyl sulfoxide resulted in progressive cell death, whereas monocytic differentiation by other differentiation inducers was only marginally affected. These results suggest that granulocytic, unlike monocytic, differentiation requires c-myb-conditioned proliferation and the activity of the protein encoded by c-fes.     

Control of normal differentiation of myeloid leukemic cells. XI. Induction of a specific requirement for cell viability and growth during the differentiation of myeloid leukemic cells

Normal hematopoietic cells require the presence of a protein (MGI) in the appropriate conditioned medium (CM) for cell viability and growth and for differentiation to mature macrophages and granulocytes. Clones of myeloid leukemic cells have been established in culture (D⁺ clones) which require CM with this protein for differentiation, but not for cell viability and growth. It has been shown that these leukemic cells can be induced by CM to again require, like normal cells, the presence of CM for cell viability and growth. Induction of this requirement, which will be referred to as RVG, occurred before the D⁺ cells differentiated to mature granulocytes. Clones of myeloid leukemic cells (D^? clones) that could not be induced to differentiate to mature cells, did not show the induction of RVG. The steroid hormones prednisolone and dexamethasone can induce some, but not all the changes associated with differentiation of D⁺ cells. Incubation with these steroids did not result in the induction of a requirement for these steroids for cell growth and viability. Studies with CM from different sources have shown, that all batches that induced RVG also induced differentiation of D⁺ cells and that both activities were inhibited after treating the CM with trypsin. It is suggested that the same protein (MGI) may be involved in both activities. Incubation of D⁺ cells with CM resulted in an increase in agglutinability by concanavalin A and this increase was maintained even in the absence of CM. This suggests, that the induction of RVG in D⁺ myeloid leukemic cells is associated with a change in the cell surface membrane.     

Differing responses of G2-related genes during differentiation of HL60 cells induced by TPA or DMSO

Differentiation leads to the cessation of cellular proliferation, but little is known about the molecular mechanisms of growth arrest. We compared the effect of two differentiation inducers, 12-o-tetradecanoyl 13-acetate (TPA) and dimethyl sulfoxide (DMSO) on both the cell-cycle and the modulation of G2-related genes in synchronized HL60 cells. TPA treatment of HL60 cells resulted in G1 arrest within 24 h. In contrast, the cell cycling of DMSO-treated cells was initially accelerated and they progressed to the second cycle before accumulating in the G1 phase. Expression of cyclin B, cdc25, wee1 and cdc2 was studied during cell cycle arrest by Northern blot hybridization. Expression of cyclin B, cdc25 and cdc2 fluctuated in association with cell cycle progression towards the G2/M phase, while wee1 expression remained constant in untreated cells. These four genes were highly expressed in TPA-treated cells for the first 12 h, but drastic down-regulation was seen at 18 h and expression became undetectable after 24 h. In contrast, no remarked changes of gene expression were seen in DMSO-treated cells. These findings suggest that cell cycle progression along with the initial process of differentiation in response to TPA differs from the response to DMSO and that the down-regulation of cdc2 expression by TPA-treated HL60 cells contributes to endorsement of G1 arrest.     

Collapse of the inner mitochondrial transmembrane potential is not required for apoptosis of HL60 cells.

Apoptotic cell death involves a series of morphological and biochemical changes orchestrated by activated proteases belonging to the caspase family. Recent studies have suggested that the activation of this process of execution is dependent upon events associated with the loss of mitochondrial inner transmembrane potential (Deltapsi(m)), as a consequence of the formation of the permeability transition (PT) pore. This has led to the proposal that mitochondrial depolarization represents a central irreversible checkpoint in the apoptotic program. Here, we present evidence that HL-60 cells undergo apoptosis in response to the cytotoxic insults of actinomycin-D, etoposide, and staurosporine without showing significant changes in Deltapsi(m). Instead, the loss of Deltapsi(m) could be detected only later in the cell death pathway. In addition, the uncoupling agent CCCP produced an early mitochondrial depolarization in HL-60s but these cells showed few signs of apoptosis up to 8 h after the insult. Furthermore, examination of these cells in response to staurosporine revealed the release of mitochondrial cytochrome c into the cytosol over time, corresponding to caspase activation irrespective of mitochondrial depolarization. In summary, our data suggest that the collapse of Deltapsi(m) as a consequence of PT is not a universal early marker for apoptosis and, moreover, it is not part of the central apoptotic machinery.     

Modification of the glyoxalase system in human HL60 promyelocytic leukaemia cells during differentiation to neutrophils in vitro

The glyoxalase system of human promyelocytic leukaemia HL60 cells was substantially modified during differentiation to neutrophils. The activity of glyoxalase I was decreased and the activity of glyoxalase II was markedly increased relative to the level in control HL60 promyelocytes. There was a decrease in the apparent maximum velocity, Vmax, of glyoxalase I, and an increase in the Vmax of glyoxalase II. The apparent Michaelis constants for both enzymes remained unchanged. The flux of intermediates metabolised via the glyoxalase system increased during differentiation, as judged by the formation of D-lactic acid, whereas the percentage of glucotriose metabolised via the glyoxalase system remained unchanged. The cellular concentrations of the glyoxalase substrates, methylglyoxal and S-D-lactoylglutathione, were markedly decreased during differentiation. The maturation of HL60 promyelocytes is associated with an increased ability to metabolise S-D-lactoylglutathione by glyoxalase II and a concomitant decrease in the mean intracellular concentrations of S-D-lactoylglutathione and methylglyoxal. The maintenance of a high concentration of S-D-lactoylglutathione in HL60 promyelocytes may be related to the status of the microtubular cytoskeleton, since S-D-lactoylglutathione potentiates the GTP-promoted assembly of microtubules.