Neural precursor cells possess multiple p53-dependent apoptotic pathways.

Neural precursor cells (NPCs) are markedly sensitive to apoptotic insults. p53-Dependent transcriptional activation of proapoptotic genes has been hypothesized to regulate NPC death in response to DNA damage. Recent studies of non-NPCs have also indicated that p53 may directly interact with Bcl-2 molecules and thereby regulate death independently of transcription. The contribution of transcription-independent p53 activation in NPC death has not been characterized. In this study, we found that apoptosis caused by chemotherapeutic agents in NPCs required p53 expression and new macromolecular synthesis. In contrast, NPC death induced by staurosporine, a broad kinase inhibitor, is regulated by p53 in the absence of macromolecular synthesis. The apoptosis effector molecules Bax and Bak, Apaf-1, and caspase-9 were shown to be downstream of p53 in both pathways. These findings indicate that p53 is in a unique position to regulate at least two distinct signaling portals that activate the intrinsic apoptotic death pathway in NPCs.

Ethanol-induced neuronal apoptosis in vivo requires BAX in the developing mouse brain.

A single episode of ethanol intoxication triggers widespread apoptotic neurodegeneration in the infant rat or mouse brain. The cell death process occurs over a 6-16 h period following ethanol administration, is accompanied by a robust display of caspase-3 enzyme activation, and meets ultrastructural criteria for apoptosis. Two apoptotic pathways (intrinsic and extrinsic) have been described, either of which may culminate in the activation of caspase-3. The intrinsic pathway is regulated by Bax and Bcl-XL and involves Bax-induced mitochondrial dysfunction and release of cytochrome c as antecedent events leading to caspase-3 activation. Activation of caspase-8 is a key event preceding caspase-3 activation in the extrinsic pathway. In the present study, following ethanol administration to infant mice, we found no change in activated caspase-8, which suggests that the extrinsic pathway is not involved in ethanol-induced apoptosis. We also found that ethanol triggers robust caspase-3 activation and apoptotic neurodegeneration in C57BL/6 wildtype mice, but induces neither phenomenon in homozygous Bax-deficient mice. Therefore, it appears that ethanol-induced neuroapoptosis is an intrinsic pathway-mediated phenomenon involving Bax-induced disruption of mitochondrial membranes and cytochrome c release as early events leading to caspase-3 activation.

p53 deficiency fails to prevent increased programmed cell death in the Bcl-X(L)-deficient nervous system.

Bcl-X(L) mice display a similar neurodevelopmental phenotype as rb, DNA ligase IV, and XRCC4 mutant embryos, suggesting that endogenous Bcl-X(L) expression may protect immature neurons from death caused by DNA damage and/or cell cycle dysregulation. To test this hypothesis, we generated bcl-x/p53 double mutants and examined neuronal cell death in vivo and in vitro. Bcl-X(L)-deficient primary telencephalic neuron cultures were highly susceptible to the apoptotic effects of cytosine arabinoside (AraC), a known genotoxic agent. In contrast, neurons lacking p53, or both Bcl-X(L) and p53, were markedly, and equivalently, resistant to AraC-induced caspase-3 activation and death in vitro indicating that Bcl-X(L) lies downstream of p53 in DNA damage-induced neuronal death. Despite the ability of p53 deficiency to protect Bcl-X(L)-deficient neurons from DNA damage-induced apoptosis in vitro, p53 deficiency had no effect on the increased caspase-3 activation and neuronal cell death observed in the developing Bcl-X(L)-deficient nervous system. These findings suggest that Bcl-X(L) expression in the developing nervous system critically regulates neuronal responsiveness to an apoptotic stimulus other than inadequate DNA repair or cell cycle abnormalities.

Chloroquine-induced neuronal cell death is p53 and Bcl-2 family-dependent but caspase-independent.

Chloroquine is a lysosomotropic agent that causes marked changes in intracellular protein processing and trafficking and extensive autophagic vacuole formation. Chloroquine may be cytotoxic and has been used as a model of lysosomal-dependent cell death. Recent studies indicate that autophagic cell death may involve Bcl-2 family members and share some features with caspase-dependent apoptotic death. To determine the molecular pathway of chloroquine-induced neuronal cell death, we examined the effects of chloroquine on primary telencephalic neuronal cultures derived from mice with targeted gene disruptions in p53, and various caspase and bcl-2 family members. In wild-type neurons, chloroquine produced concentration- and time-dependent accumulation of autophagosomes, caspase-3 activation, and cell death. Cell death was inhibited by 3-methyladenine, an inhibitor of autophagic vacuole formation, but not by Boc-Asp-FMK (BAF), a broad caspase inhibitor. Targeted gene disruptions of p53 and bax inhibited and bcl-x potentiated chloroquine-induced neuron death. Caspase-9- and caspase-3-deficient neurons were not protected from chloroquine cytotoxicity. These studies indicate that chloroquine activates a regulated cell death pathway that partially overlaps with the apoptotic cascade.

Neural precursor cell apoptosis and glial tumorigenesis following transplacental ethyl-nitrosourea exposure.

Neural precursor cells (NPCs) populate the embryonic ventricular zone and persist in the subependymal zone of the adult brain. We hypothesized that hereditary and/or acquired mutations in apoptosis-associated genes, such as p53 and caspases, may protect NPCs from DNA damage-induced death and predispose them to subsequent neoplastic transformation. To test this hypothesis, we exposed NPCs from wild-type and targeted gene-disrupted mouse embryos (p53, caspase-9, caspase-3, and bax mutants) to ethyl-nitrosourea (ENU), a known DNA mutagen and neural carcinogen, and measured NPC viability. We found that ENU produced caspase-3 activation and apoptotic NPC death 6-24 h after administration both in vivo and in vitro. This effect was critically dependent on p53 and caspase-9 expression. The long-term effect of intrauterine ENU exposure was examined in control and p53-deficient mice. High grade glial tumors were found in 60% of p53(-/-) young adult mice exposed to ENU on gestational day 12.5 but not in p53(+/-) or p53(+/+) littermates or in untreated p53-deficient mice. All the tumors were located supratentorially and possessed strong immunoreactivity for glial fibrillary acidic protein and the anti-apoptotic molecule Bcl-X(L). These results suggest that intrauterine exposure of NPCs to certain DNA damaging agents may synergistically interact with specific genetic abnormalities (e.g. p53 deficiency) to produce glial neoplasms in the adult brain.

Distal regulatory elements control MRF4 gene expression in early and late myogenic cell populations.

MRF4 is a muscle-specific transcription factor that belongs to a family of basic helix-loop-helix proteins known as the myogenic regulatory factors (MRFs). In vitro studies have shown that expression of the MRF4 gene is controlled by a proximal promoter element (-336 to +71) that binds the muscle-specific transcription factors MEF2 and myogenin to activate transcription. To examine further the regulatory elements necessary for endogenous MRF4 gene expression during development, transgenic mice were generated that contained either a proximal MRF4 promoter-LacZ reporter gene (-336 MRF4-nLacZ) or a MRF4-LacZ reporter gene containing 8.5 kb of 5' flanking sequence (-8500 MRF4-nLacZ). Characterization of individual transgenic mouse lines throughout development revealed that expression of both transgenes is restricted to skeletal muscle tissue. However, unlike previous in vitro data, the proximal promoter transgene exhibits only limited transcriptional activity at all developmental time points, whereas the -8500 MRF4-nLacZ lines fully recapitulate the later developmental expression patterns and exhibit transcription in myotomal cells during somitic differentiation. Tissue culture analysis of myogenic cells isolated from E12.5, E16.5, and adults confirmed that the -8500 MRF4-nLacZ transgene is expressed in greater than 90% of the myotubes for all myogenic populations. These results indicate that 8.5 kb of MRF4 5' flanking sequence contains all the regulatory elements necessary for late MRF4 expression and that at least some of these elements lie upstream of the -336 proximal promoter. It is also likely that distant upstream regulatory sequences control early somitic MRF4 expression. These findings, coupled with previous in vitro studies, suggest that the early and late developmental expression patterns of the MRF4 gene are controlled by distinct sets of regulatory elements.

Maturation of the acetylcholine receptor in skeletal muscle: regulation of the AChR gamma-to-epsilon switch.

During the development of the mammalian neuromuscular junction, acetylcholine receptors (AChRs) become localized to the postsynaptic muscle membrane. As this process nears completion, the fetal form of the receptor, containing a gamma subunit (composition alpha 2 beta gamma delta) is gradually replaced by an epsilon subunit-containing adult form (alpha 2 beta epsilon delta). To understand how this transition is controlled, we compared the expression and regulation of the AChR gamma and epsilon subunits in developing, adult, and cultured muscles. Immunostaining with subunit-specific antibodies showed that replacement of gamma subunit- by epsilon subunit-containing AChRs occurs largely during the first postnatal week in fast-twitch muscles, and occurs homogeneously throughout individual endplates. In the slow-twitch soleus, however, this transition is delayed, and in the multiply innervated slow fibers of extraocular muscle, gamma subunit expression persists into adulthood. The transcriptional bases of the AChR subunit transition, and of these intermuscular variations, were demonstrated in mice bearing transgenes containing promoter elements from the AChR gamma and epsilon subunit genes, each coupled to a nuclear-localized beta-galactosidase (nlacZ) reporter. We show that transgene expression is stimulated by the nerve-derived inducer of AChR expression, ARIA, in myotubes cultured from gamma-nlacZ as well as epsilon-nlacZ mice. However, the expression of gamma-nlacZ, but not epsilon-nlacZ, is increased by treatment of myotubes with TTX, and the ARIA sensitivity of gamma-nlacZ is dependent on the electrical state of the myotube. Thus, the promoters of the gamma and epsilon subunit genes may integrate ARIA- and activity-dependent signals in different ways to generate their complementary patterns of expression.

Overexpression of myogenin in muscles of transgenic mice: interaction with Id-1, negative crossregulation of myogenic factors, and induction of extrasynaptic acetylcholine receptor expression.

To investigate the role of myogenin in regulating acetylcholine receptor expression in adult muscle, this muscle-specific basic helix-loop-helix transcription factor was overexpressed in transgenic mice by using regulatory elements conferring strong expression confined to differentiated postmitotic muscle fibers. Many of the transgenic mice died during the first postnatal week, but those that survived into adulthood displayed normal muscle histology, gross morphology, and motor behavior. The mRNA levels of all five acetylcholine receptor subunits (alpha, beta, gamma, delta, and epsilon) were, however, elevated. Also, the level of receptor protein was increased and high levels of receptors were present throughout the extrasynaptic surface membrane of the muscle fibers. Thus, elevated levels of myogenin are apparently sufficient to induce acetylcholine supersensitivity in normally innervated muscle of adult mice. The high neonatal mortality rate of the mice overexpressing myogenin hindered the propagation of a stable line. In an attempt to increase survival, myogenin overexpressers were mated with a line of transgenic mice overexpressing Id-1, a negative regulator that interacts with the basic helix-loop-helix family of transcription factors. The Id-1 transgene apparently worked as a second site suppressor and abolished the high rate of neonatal mortality. This effect indicates that Id-1 and myogenin interact directly or indirectly in these animals. Further study indicated that myogenin overexpression had no effect on the level of endogenous myogenin mRNA, while the levels of myoD and MRF4 mRNAs were reduced. Overexpression of the negative regulator Id-1 increased the mRNA levels of all the myogenic factors. These findings are consistent with a hypothesis suggesting that myogenic factors are influenced by mechanisms that maintain cellular homeostasis.