Search

Murine embryonic stem (ES) cells are maintained in an undifferentiated state when cultured in medium conditioned by Buffalo rat liver (BRL) cells. BRL conditioned medium (CM) contains a differentiation inhibitory activity (DIA) that is synonymous with leukemia inhibitory factor (LIF). ES cells in monolayer culture can be induced to differentiate by addition of all-trans retinoic acid (RA) to the BRL CM, when they mainly form cells resembling parietal endoderm, or by culture in medium not conditioned by BRL cells. ES cells thus deprived of LIF/DIA differentiate spontaneously to a cell type that expresses Brachyury (T), a marker of early mesoderm. Northern blot analyses have shown previously that transcripts for transforming growth factor beta 1 (TGF-beta 1) are detected in undifferentiated cells while transcripts for TGF-beta 2 and TGF-beta 3 only become detectable after differentiation. We have now determined levels of TGF-beta protein in CM and in the extracellular matrix (ECM) and have used neutralizing antibodies specific for TGF-beta 1 and TGF-beta 2 that do not react with recombinant human TGF-beta 3 to determine the isoform secreted. Using the growth inhibition of mink lung CCL64 cells as a bioassay for TGF-beta activity, we demonstrate that undifferentiated ES cells secrete latent TGF-beta 1 into the medium but no activity is found in their ECM. Cells induced to differentiate with RA contain TGF-beta 2 in both active and latent forms in their CM. Likewise their ECM contains TGF-beta 2 as the sole isoform. ES cells deprived of LIF/DIA secrete both TGF-beta 1 and TGF-beta 2 isoforms in their CM but TGF-beta-like activity remains after addition of neutralizing antibodies for TGF-beta 1 and TGF-beta 2. This active TGF beta is the major component of the TGF-beta activity in this CM. By contrast, ECM from LIF/DIA deprived cells contains only the TGF-beta 1 and beta 2 isoforms. The remaining activity in CM correlates with high expression of TGF-beta 3 by Northern blot analysis in these cells. We speculate that TGF-beta 3 is secreted by these cells and may be activated more efficiently and/or in a different manner to TGF-beta 1 and TGF-beta 2, since it is present in CM only in its active form.

Cation transport and membrane potential were studied during the cell cycle of neuroblastoma cells (clone Neuro-2A) to investigate the role of these parameters in growth regulation. The cells were synchronized by selective detachment of mitotic cells. The membrane potential and intracellular K+ activity were measured with conventional and K+-selective microelectrodes respectively. Both the membrane potential and K+ activity were high in mitosis, decreased to half maximal in G1 phase, and rose again during S phase. K+ efflux across the plasma membrane was studied with 42K+ as a radioactive tracer using a washing method for cells grown in monolayer and a continuous efflux method for mitotic cells in suspension. The intracellular K+ content and unidirectional K+ efflux rate obtained from these measurements showed modulations during the cell cycle similar to those of the membrane potential. Using equations of electrodiffusion theory the membrane permeabilities to K+ and Na+ were calculated. These permeabilities were high in mitosis, decreased rapidly in G1 phase and increased during S phase, followed by a transient decrease in G2 phase. A rapid increase was observed between G2 phase and the next mitosis. A similar pattern was obtained for the K+ conductance. K+ resistance changes during the cell cycle were similar to changes in the specific membrane resistance, measured by microelectrodes, except for the early cell cycle phases (mitosis and G1). These studies clearly demonstrate large modulations of the passive membrane permeability properties during the cell cycle. These modulations can be correlated with physicochemical membrane variations during the cell cycle, such as membrane fluidity and lateral mobility of lipids.

Rat pheochromocytoma cells (clone PC12) possess functional surface receptors for both nerve growth factor (NGF) and epidermal growth factor (EGF). PC12 cells respond to NGF as well as to dibutyryl cyclic AMP (dbcAMP) by arrest of cell proliferation and initiation of morphological differentiation, while EGF acts as a mitogen. Exposure of PC12 cells to NGF for several days resulted in a complete loss of rapid EGF responses, such as membrane ruffling and activation of active K+ transport. EGF binding studies revealed that this loss of EGF responses was due to an almost complete reduction of the number of EGF binding sites. In contrast, exposure of PC12 cells to dbcAMP for 2 days did not affect the rapid EGF responses, despite the morphological differentiation. Moreover, EGF binding studies demonstrated a twofold increase in the number of high-affinity binding sites and a small increase in the number of low-affinity sites. In addition, exposure of the cells to dbcAMP caused a twofold increase of EGF-receptor phosphotyrosine kinase activity. These results indicate that neither EGF-binding or the presence of EGF receptors nor the rapid EGF responses are sufficient for persistent proliferation, on one hand, or sufficient to avoid morphological differentiation, on the other.

Rat pheochromocytoma cells (clone PC12) display cell surface receptors for both nerve growth factor (NGF) and epidermal growth factor (EGF) and therefore provide a useful model system with which to study the role of these receptors in the regulation of proliferation and differentiation. In this paper PC12 cells are demonstrated to possess two classes of EGF receptors, a high-affinity class with 7,600 sites per cell and an apparent dissociation constant (Kd) of 0.05 nM, and a low-affinity class with 62,000 sites per cell and a Kd of 14.1 nM. These findings are contrary to literature data (Huff et al., 1981; Vale and Shooter, 1983) but can be explained in part by differences in experimental conditions. Binding studies at 37 degrees C compared with room temperature demonstrated similar affinities of both classes, but during prolonged incubation at 37 degrees C, the binding capacities of both classes decreased. Furthermore the high-affinity class was sensitive to lectins, such as concanavalin A (Con A), and to the tumor-promoting phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA). Both compounds caused a decrease of the affinity of the high-affinity class without affecting the low-affinity class. At high concentrations of Con A or TPA, a decrease of the apparent number of binding sites of the low-affinity class was also observed. The similarities between the characteristics of EGF binding and NGF binding in PC12 cells are striking and will be discussed.

Functional and optimal activities of the (Na+-K+)ATPase, as determined by ouabain-sensitive K+ influx in intact cells and ATP hydrolysis in cell homogenates respectively, have been measured during the cell cycle of neuroblastoma (clone Neuro-2A) cells. The cells were synchronized by selective detachment of mitotic cells. The ouabain-sensitive K+ influx decreased more than fourfold from 1.62 +/- 0.11 nmoles/min/10(6) cells to 0.36 +/- 0.25 nmoles/min/10(6) cells on passing from mitosis to early G1 phase. On entry into S phase a transient sixfold increase to 2.07 +/- 0.30 nmoles/min/10(6) cells was observed, followed by a rapid decline, after which the active K+ influx rose again steadily from 1.03 +/- 0.25 nmoles/min/10(6) cells in early S phase to 2.10 +/- 0.92 nmoles/min/10(6) cells just prior to the next mitosis. The ouabain-insensitive component rose linearly through the cycle in the same manner as the protein content/cell. Combining total K+ influx values with efflux data obtained previously showed that net loss of K+ occurred with transition from mitosis to G1 phase while net accumulation occurred with entry into S. Throughout mid-S phase net K+ flux was virtually zero, but a large net influx occurred again just before the next mitosis. The (Na+-K+)ATPase activity measured in cell homogenates decreased rapidly from mitosis to G1 phase and increased steadily throughout S phase, but the transient activation on entry into S phase was not observed. Complete inhibition of the (Na+-K+)ATPase mediated K+ influx by ouabain (5 mM) prevents the cells from entering S phase, while partial inhibition by lower concentrations of ouabain (0.2 and 0.5 mM; km = 0.17 mM) causes partial blockage in G1 and, to a lesser extent, a reduced rate of progression through the rest of the cell cycle. We conclude that the transient increase in (Na+-K+)ATPase mediated K+ influx at the G1/S transition is a prerequisite for entry into S phase, while maintenance of adequate levels of K+ influx is necessary for normal rate of progression through the rest of the cell cycle.