Functional hepatocyte-like cells derived from mouse embryonic stem cells: a novel in vitro hepatotoxicity model for drug screening.

The aim of the present study was to differentiate mouse embryonic stem (mES) cells into a high percentage of hepatocyte-like cells, and to demonstrate their utility as an in vitro hepatotoxicity model. We were able to differentiate 80-90% of mES cells using optimized hepatocyte differentiation medium. These differentiated cells showed typical hepatocyte morphology, expressed hepatic specific genes as shown by RT-PCR and displayed antibody detectable expression of markers specific for hepatic maturation. These hepatocyte-like cells also demonstrated evidence of glycogen storage. These cells when exposed to CCl4, a commonly used hepatotoxicant, showed an elevation of liver function enzymes, SGOT, SGPT and LDH, indicating hepatic damage. Further, this increase was prevented by pre-treatment with N-acetylcysteine, a known anti-oxidant. Thus we propose that the hepatocyte-like cells derived by the present method may prove to be useful as an in vitro model of hepatotoxicity, thereby providing a novel and promising alternative for obtaining large numbers of functional hepatocyte-like cells for in vitro drug metabolism and hepatotoxicity screening of potential drug candidates.

2'-Deoxyadenosine causes cell death in embryonic chicken sympathetic ganglia and brain.

Previous work has shown that nucleosides produce apoptosis in sympathetic ganglion (SG) cells in vitro. The present study examined the effects of nucleosides on the development of the chick embryo in vivo with special attention to the SG and the optic tectum of the central nervous system. In the presence of an adenosine deaminase inhibitor, adenosine and 2'-deoxyadenosine (2'-dAdo) produced different toxicity patterns: both adenosine and 2'-dAdo were toxic to E3 embryos, but only 2'-dAdo was toxic at later stages (E6 1/2, E11). Dosage experiments on E6 1/2 embryos showed that adenosine was less toxic than 2'-dAdo and that 2'-dAdo in sublethal doses was teratogenic. We also examined the effects of 2'-dAdo on embryonic chicken SG and optic tectum in vivo to determine whether sublethal doses of 2'-dAdo produced cell death in these centers on E6 1/2 and 10. In the E6 1/2 SG, 2'-dAdo produced significant neuron loss (83%) and a decrease in SG volume (65%); however, at E10, there was only minor cell loss (7%) and no significant change in SG volume. In the optic tectum at E6 1/2, cell loss was confined mainly to the tectal ventricular zone, but there was little sign of cell loss in this organ at E10. Since cell production is vigorous in the SG and optic tectum at E6 1/2 but relatively low at E10, 2'-dAdo appears to work by stopping cell proliferation. The ineffectiveness of 2'-dAdo at E10 may result from the lethality of 2'-dAdo to the embryo at low concentrations (30 microM) in vivo, well below the apoptosis-inducing concentrations employed in vitro (100-300 microM). These data extend previous findings showing that purine and pyrimidine metabolism plays an important role in development.

Pituitary adenylyl cyclase-activating polypeptide and nerve growth factor use the proteasome to rescue nerve growth factor-deprived sympathetic neurons cultured from chick embryos.

Removal of nerve growth factor (NGF) from sympathetic neurons initiates a neuronal death program and apoptosis. We show that pituitary adenylyl cyclase-activating polypeptide (PACAP) prevents apoptosis in NGF-deprived sympathetic neurons. PACAP (100 nM) added to culture medium at the time of plating failed to support neuronal survival. However, in neurons grown for 2 days with NGF and then deprived of NGF, PACAP prevented cell death for the next 24-48 h. Uptake of [3H]norepinephrine ([3H]NE) was used as an index of survival and decreased >50% in NGF-deprived cultures within 24 h. PACAP (1-100 nM) restored [3H]NE uptake to 92 +/- 8% of that of NGF-supported controls. Depolarization-induced [3H]NE release in neurons rescued by PACAP was the same as that in NGF-supported neurons. PACAP rescue was not mimicked by forskolin or 8-bromo-cyclic AMP and was not blocked by the protein kinase A inhibitor Rp-adenosine 3',5'-cyclic monophosphothioate. Mobilization of phosphatidylinositol by muscarine failed to support NGF-deprived neurons. Thus, PACAP may use novel signaling to promote survival of sympathetic neurons. The apoptosis-associated caspase CPP32 activity increased approximately fourfold during 6 h of NGF withdrawal (145 +/- 40 versus 38 +/- 17 nmol of substrate cleaved/min/mg of protein) and returned to even below the control level in NGF-deprived, PACAP-rescued cultures (14 +/- 7 nmol/min/mg of protein). Readdition of NGF or PACAP to NGF-deprived cultures reversed CPP32 activation, and this was blocked by lactacystin, a potent and specific inhibitor of the 20S proteasome, suggesting that NGF and PACAP target CPP32 for destruction by the proteasome. As PACAP is a preganglionic neurotransmitter in autonomic ganglia, we propose a novel function for this transmitter as an apoptotic rescuer of sympathetic neurons when the supply of NGF is compromised.

Adenosine protects chick embryonic sympathetic neurons from apoptotic death by 2'-deoxyadenosine--importance of ATP in apoptosis.

Our past work on nucleoside toxicity in sympathetic neurons has clearly revealed that adenosine and 2'-deoxyadenosine (dAdo) have different mechanisms of action in inducing apoptotic death. For example, adenosine is toxic to neurons only during early phase of growth whereas dAdo kills even mature neurons. In this study, we hypothesize that dAdo-induced apoptosis is initiated when ATP concentration of sympathetic neurons decreases below a critical level. To prove our hypothesis we used adenosine as a tool to replenish ATP levels of sympathetic neurons. We demonstrate that dAdo toxicity in mature sympathetic neurons was fully prevented by adenosine treatment. Furthermore, we demonstrate that depletion of ATP caused by dAdo was prevented by pretreatment with adenosine. These data suggest that intracellular accumulation of adenosine could play a neuroprotective role in preventing death associated with reduction in neuronal ATP concentration.

Adenosine induces apoptosis by inhibiting mRNA and protein synthesis in chick embryonic sympathetic neurons.

Our previous work has established that adenosine is toxic to chick embryonic sympathetic neurons and kills freshly plated neurons by a process of apoptosis. Although the exact mechanism remains unknown, we found that phosphorylation of adenosine was essential to the toxicity. Using markers for RNA ([3H]uridine) and protein ([35S]methionine) synthesis we demonstrate here that in freshly plated sympathetic neurons adenosine inhibits RNA and protein synthesis by about 50%. The inhibitory effects of adenosine on RNA and protein synthesis, and increased ATP synthesis were blocked by adenosine kinase inhibitor, suggesting that phosphorylated products are responsible for inhibition of RNA and protein synthesis and cell death. Adenosine-induced inhibition of RNA and protein synthesis in neuronal cells provides a new role for adenosine in the regulation of cell function.

2'-Deoxyadenosine selectively kills nonneuronal cells without affecting survival and growth of chick dorsal root ganglion neurons.

Recently, we have demonstrated that adenosine and 2'-deoxyadenosine are toxic to embryonic sympathetic neurons and proposed that purine and pyrimidine metabolism may play a critical role in the growth and development of sympathetic neurons. To extend this hypothesis further, we examined the effects of these nucleosides on two other neuronal populations in the chick embryo, sensory dorsal root ganglion neurons and parasympathetic ciliary ganglion neurons. Now, we show that 2'-deoxyadenosine and adenosine have no visible adverse effect on the viability of either sensory or parasympathetic neurons. Instead, 2'-deoxyadenosine proved to be highly toxic to the nonneuronal cells. The toxic effects of 2'-deoxyadenosine were markedly enhanced by inhibition of adenosine deaminase. In contrast, adenosine was much less toxic to nonneuronal cells than 2'-deoxyadenosine and its effect was not potentiated by inhibition of adenosine deaminase. Priming of pyrimidine pools by exogenous uridine and the specific inhibitor of the nucleoside transporter, nitrobenzylthioinosine, did not protect nonneuronal cells from 2'-deoxyadenosine toxicity. Since phosphorylation of internalized nucleosides was a key step in the initiation of toxicity in sympathetic neurons, adenosine kinase activity was compared in sensory and sympathetic neuronal cultures. The adenosine kinase activity in dorsal root ganglion cultures was only 20% of that in sympathetic ganglion cultures. Furthermore, inhibition of phosphorylation by blocking 2'-deoxyadenosine kinase with iodotubercidin and 5'-amino-5'-deoxyadenosine had no protective effect against 2'-deoxyadenosine toxicity. [3H]-thymidine incorporation was inhibited over 90% by 2'-deoxyadenosine as early as 6 h following its addition and for up to 4 days, suggesting inhibition of proliferation of nonneuronal cells by 2'-deoxyadenosine. The nucleoside was also able to wipe out already well established nonneuronal cells, leaving behind an enriched population of sensory neurons. The selective vulnerability of nonneuronal cells to 2'-deoxyadenosine offers a convenient and effective tool for removing nonneuronal cells from neuronal cultures as well as providing a new model for studying the mechanisms of nucleoside toxicity.

Chronic activation of protein kinase C by phorbol ester reduces calcium channel expression in chick sympathetic neurons.

Chronic activation of protein kinase C (PKC) has been implicated in regulation of Ca2+ entry responsible for normal development of transmitter properties in cultured sympathetic neurons. The idea that PKC alters the expression of Ca2+ channels was tested using phorbol 12,13-dibutyrate (PDB) which activates PKC and also supports survival of chick sympathetic neurons in the absence of nerve growth factor (NGF). Whole cell voltage-clamp showed that neurons supported by PDB for 2 days had significantly lower Ca2+ current density (0.243 +/- 0.025 pA/microm2) than those supported by NGF (0.356 +/- 0.033 pA/microm2). [125I]omega-Conotoxin GVIA binding showed that PDB-supported neurons had significantly lower maximum binding (617 +/- 223 fmol/mg protein) compared with those supported by NGF (1099 +/- 192 fmol/mg protein). These results support the conclusion that chronic activation of PKC limits the expression of N-type Ca2+ channels. A reduction in Ca2+ channel number is consistent with, and could account for the mature type Ca2+ handling and transmitter release properties seen in sympathetic neuro-effector preparations, sympathetic neurons co-cultured with their targets, and neurons supported by PDB.

2'-deoxyadenosine induces apoptosis in rat chromaffin cells.

We show here that 2'-deoxyadenosine (2'-dAdo) but not adenosine was toxic to chromaffin cells of 3-4-week-old rat adrenal glands. More than 75% of the cells plated in culture gradually died over a 3-day period in the presence of 100 microM 2'-dAdo plus 3 microM deoxycoformycin (DCF). Morphological observations together with bisbenzimide staining and terminal deoxynucleotidyl transferase-mediated nick and labeling showed membrane blebbing, shrinkage of cell bodies, chromatin condensation, and DNA fragmentation, suggesting apoptosis-like cell death by 2'-dAdo. Lethal effects of 2'-dAdo were potentiated by DCF, a drug that inhibits adenosine deaminase. 2'-dAdo-prompted cell death was not prevented by inhibitors of nucleoside transporter (3 microM dilazep or 1 microM nitrobenzylthioinosine), precursors of pyrimidine nucleotide biosynthesis (300 microM uridine or 100 microM 2'-deoxycytidine), or 5 mM nicotinamide. Cells incubated with 2'-dAdo (100 and 300 microM) showed a three- and ninefold, respectively, increase in content of dATP, a product known to be an inhibitor of ribonucleotide reductase, an enzyme essential for DNA synthesis. Formation of dATP was completely prevented by iodotubercidin (ITu), a drug that inhibits phosphorylation of 2'-dAdo to dATP by nucleoside kinase. It is interesting that nanomolar concentrations of ITu also completely protected chromaffin cells from 2'-dAdo lethality. Our study demonstrates for the first time that mammalian adrenal chromaffin cells undergo apoptotic cell death by a natural nucleoside and suggests that this model could be used to study apoptosis and cell function.

Quantitative analysis of similarities and differences in neurotoxicities caused by adenosine and 2'-deoxyadenosine in sympathetic neurons.

These experiments characterize the nucleoside transport and quantify the neurotoxicity of adenosine and 2'-deoxyadenosine (dAdo) in chick sympathetic neurons. We show that [3H]adenosine transport was sensitive to low temperature, specific inhibitors of nucleoside transport, and an excess concentration of adenosine. However, many of these treatments had a marginal effect on [3H]dAdo transport. Total retention of [3H]dAdo over short and long periods was approximately 10 times less than that of [3H]adenosine. These data suggest that adenosine and dAdo enter sympathetic neurons by different routes. Uptake of (3H]norepinephrine ([3H]NE) decreased in neurons damaged by nucleosides and increased to control levels when neurons were protected by various agents against adenosine or dAdo toxicity. These results indicate that [3H]NE uptake serves as a quantitative index of toxicity by the nucleosides. Using this approach we demonstrate that phosphorylation of both nucleosides is essential for their lethal action. For example, iodotubercidin prevented nucleoside-induced neuronal death, but the effect was much more pronounced in the case of dAdo toxicity (IC50 of 0.83 +/- 0.4 vs. 30 +/- 1.6 nM). Another kinase inhibitor, 5'-amino 5'-deoxyadenosine, was effective in protecting neurons against dAdo but had no effect against adenosine toxicity. These results suggest that specific kinases are associated with the phosphorylation of adenosine and dAdo in sympathetic neurons to produce toxic metabolic products. Finally, neurons were susceptible to dAdo toxicity from the time of plating to 4 weeks in culture but were resistant to adenosine toxicity 8 h after plating. In conclusion, our results highlight major differences in the mechanism of neurotoxicity by adenosine and dAdo and provide insights for identification of biochemical pathways leading to neuronal death.

Importance of protein kinase C for normal development of transmitter release properties in embryonic chick sympathetic neurons in culture.

The hypothesis that multiple trophic inputs are essential for normal development of transmitter release properties in sympathetic neurons was tested using two supportive agents (excess KCl and phorbol 12,13-dibutyrate which produce marked activation of protein kinase C and also support survival of chick sympathetic neurons in culture) in addition to nerve growth factor, ciliary neurotrophic factor and neurotrophin-3. Basal and electrically evoked (10 pulses at 1 Hz and 10 Hz) release of [(3)H]norepinephrine from neurons supported by nerve growth factor was very high (1.5 to 2% of total [(3)H]norepinephrine content) and relatively insensitive to facilitation by tetraethylammonium as compared to release in neuroeffector organs, and the frequency-release response was negative. In K+-supported neurons, basal [(3)H]norepinephrine release was almost four-fold lower, evoked release was four- to eight-fold lower, the frequency response was flat to positive, and tetraethylammonium increased evoked release up to four-fold. Inclusion of nerve growth factor in culture did not modify the effects of K+ on basal or evoked release, and nerve growth factor plus ciliary neurotrophic factor and/or neurotrophin-3 did not produce the changes observed in K+-supported neurons. Neurons supported by phorbol ester had a low background release, low evoked release, a positive frequency-release response, and 10- to 30-fold facilitation by tetraethylammonium of release evoked by 1 Hz or 1 pulse stimulation. Thus, physiological and pharmacological behavior of transmitter release of sympathetic neurons supported by excess KCl or phorbol ester was very similar to their counterparts growing in the body. Neurons supported by nerve growth factor showed an immediate rise in stimulated [Ca(2)+]i that was three- to five-fold above basal levels with either 1 Hz or 10 Hz stimulation. However, in phorbol supported neurons, [Ca(2)+]i rose gradually to about 1.5 times basal levels during 1 Hz stimulation and increased further with 10Hz stimulation. Tetraethylammonium had little effect on stimulated [Ca(2)+]i in nerve growth factor-supported neurons, but greatly facilitated the stimulated rise in [Ca(2)+]i in phorbol-supporte neurons. The data show that multiple trophic inputs distinct from nerve growth factor, neurotrophin-beta or ciliary neurotrophic factor are required for normal physiological function of sympathetic neurons.

Morphological and transmitter release properties are changed when sympathetic neurons are cultured in low Ca2+ culture medium.

The stimulated elevation of [Ca2+]i can either promote neuronal survival or lead to Ca(2+)-mediated neurotoxicity. Similarly, growth cone mobility and neurite outgrowth may be promoted or arrested by elevated [Ca2+]i. We examined survival, development and transmitter release properties of chick sympathetic neurons maintained in culture medium containing varying concentrations of Ca2+. Neurons maintained in medium with no added Ca2+ or as low as 0.1 mM external Ca2+ show a dramatic change in growth and development compared to neurons kept in 1-2 mM Ca(2+)-containing medium. Furthermore, neurons in Ca(2+)-free medium (+ 100 microM EGTA) survived up to 24 h and, following change to 0.1 mM Ca2+, grew neurites and survived for several weeks. Neurons grown in high Ca2+ medium (0.6-2 mM) exhibited thick neurites, aggregated cell bodies, and neurites began to detach after six to eight days in culture. Neurons in low Ca2+ medium (no added Ca2+ to 0.3 mM) grew as single cells with extensive, thin branching neurites, remained firmly attached to the substrate and survived for several weeks. Neurons initially plated in 0.1 mM Ca2+ (or 2 mM Ca2+) medium and switched over to 2 mM (or 0.1 mM) Ca2+ medium after two days acquired the characteristic morphology of high (and low) Ca2+ medium over the next six days, demonstrating the plasticity of effects of external Ca2+. The above characteristic changes in growth of sympathetic neurons in low Ca2+ medium occurred when neurons were supported by 35 mM KCl or 30 nM phorbol 12,13-dibutyrate instead of nerve growth factor. The uptake and retention of tritiated norepinephrine in neurons grown in low or high Ca2+ medium were similar. However, basal release of [3H]norepinephrine in neurons maintained in low Ca2+ medium was one-third of that in neurons kept in high Ca2+ medium. Furthermore, electrically stimulated (10 pulses at 1 or 10 Hz) [3H]norepinephrine release from neurons grown in high Ca2+ had a high fractional release (> 1%) which did not change during six days in culture. On the other hand, fractional release in neurons grown in low Ca2+ medium for six to 10 days decreased about 50% and 75%, respectively, compared to release after two days in culture. The resulting low fractional release (< 0.5%) is characteristic of sympathetic neurons in neuroeffector organs.(ABSTRACT TRUNCATED AT 400 WORDS)

Deoxynucleoside induces neuronal apoptosis independent of neurotrophic factors.

Postmitotic sympathetic neurons are known to undergo a programmed cell death (apoptosis) when they are deprived of nerve growth factor (NGF) or treated with arabinofuranosyl nucleoside antimetabolites. Here we report the existence of a biochemical mechanism for the induction of neuronal death by an endogenous nucleoside in the presence of NGF. In support of such a mechanism we show that 2-deoxyadenosine (dAdo) induces apoptosis in chick embryonic sympathetic neurons supported in culture by NGF, excess K+, phorbol 12,13-dibutyrate, or forskolin. Neuronal death was related to a dramatic increase in the dATP content of sympathetic neurons exposed to dAdo (34.96 +/- 5.98 versus 0.75 +/- 0.16 pmol/micrograms protein in untreated controls, n = 9), implicating dATP in the toxicity. Supportive evidence for a central role of dATP was gained by inhibition of kinases necessary for phosphorylation of dAdo. 5'-Iodotubercidin in nanomolar concentrations completely and dose-dependently inhibited formation of dATP and also protected against toxicity of submillimolar concentrations of dAdo in sympathetic neurons. Although some of these actions of dAdo were remarkably similar to those reported for human lymphoid cells, several were uniquely different. For example, [3H]dAdo was not transported into neurons by the nucleoside transporter, and therefore inhibition of the transporter (dilazep, nitrobenzylthioinosine) did not prevent neurotoxicity by dAdo. Precursors of pyrimidine synthesis (2'-deoxycytidine, uridine) or NAD+ synthesis (nicotinamide) were ineffective in protecting sympathetic neurons against dAdo toxicity. Finally, inhibition of adenosine deaminase by deoxycoformycin or erythro-9-(2-hydroxy-3-nonyl) adenine did not potentiate the toxic effects of dAdo. Our results provide evidence for the first time that neuronal cells are as susceptible to nucleoside lethality as human lymphocytes are, and provide a new model to study the salvage pathway of deoxyribonucleosides in controlling neuronal populations through programmed cell death.