Astrocytes exert a pro-apoptotic effect on neurons in postnatal hippocampal cultures.

We investigated conditions that promote basal and activity-dependent neuronal apoptosis in postnatal rat hippocampal cultures. Low-density mixed cultures of astrocytes and neurons exhibited lower sensitivity than high-density cultures to basal neuronal death and activity-sensitive neuronal death, induced with glutamate receptor blockers, sodium channel blockers, or calcium channel blockers. Although elevations of [Ca(2+)](i) protect neurons from apoptosis, low-density microcultures and mass cultures exhibited only minor differences in resting [Ca(2+)](i) and Ca(2+) current density, suggesting that these variables are unlikely to explain differences in susceptibility. Astrocytes, rather than neurons, were implicated in the neuronal loss. Several candidate molecules implicated in other astrocyte-dependent neurotoxicity models were excluded, but heat inactivation experiments suggested that a heat-labile factor is critically involved. In sum, our results suggest the surprising result that astrocytes can be negative modulators of neuronal survival during development and when the immature nervous system is challenged with drugs that dampen electrical excitability.

Induction of BIM, a proapoptotic BH3-only BCL-2 family member, is critical for neuronal apoptosis.

Sympathetic neuronal death induced by nerve growth factor (NGF) deprivation requires the macromolecular synthesis-dependent translocation of BAX from the cytosol to mitochondria and its subsequent integration into the mitochondrial outer membrane, followed by BAX-mediated cytochrome c (cyt c) release. The gene products triggering this process remain unknown. Here, we report that BIM, a member of the BH3-only proapoptotic subfamily of the BCL-2 protein family, is one such molecule. NGF withdrawal induced expression of BIM(EL), an integral mitochondrial membrane protein that functions upstream of (or in parallel with) the BAX/BCL-2 and caspase checkpoints. Bim deletion conferred protection against developmental and induced neuronal apoptosis in both central and peripheral populations, but only transiently, suggesting that BIM--and perhaps other BH3-only proteins--serve partially redundant functions upstream of BAX-mediated cyt c release.

Analysis of a novel mechanism of neuronal toxicity produced by an apolipoprotein E-derived peptide.

The apolipoprotein E (apoE)-derived peptide (141-155)2 has a neurotoxic effect, implying that apoE itself could be a source of toxicity in Alzheimer's disease brain. We characterized the toxicity of this peptide on superior cervical ganglion (SCG) neurons and compared the death with the apoptotic death that occurs after nerve growth factor (NGF) deprivation in these cells. A dose of 10 microM apoE (141-155)2 resulted in the death of approximately 50% of the neurons within 24 h. Nuclear condensation and DNA fragmentation preceded the death. However, most inhibitors of NGF deprivation-induced death, including the caspase inhibitor Boc-aspartyl(O-methyl)fluoromethyl ketone and genetic deletion of bax-/-, had no effect on the toxicity. Inclusion of depolarizing levels of potassium did block the toxicity. Receptor-associated peptide (RAP), an antagonist for apoE receptors, did not protect cells in either SCG or hippocampal cultures. In addition, RAP had no effect on internalization of the apoE peptide. These data support the observation that apoE (141-155)2 is neurotoxic but suggest that the neurotoxicity is distinct from classical apoptosis or necrosis. Furthermore, these results indicate that the toxic effect may occur independently of members of the low-density lipoprotein receptor gene family.

Generation of neuronal intranuclear inclusions by polyglutamine-GFP: analysis of inclusion clearance and toxicity as a function of polyglutamine length.

Recent evidence suggests that, in huntingtin and many other proteins, polyglutamine repeats are a toxic stimulus in neurodegenerative diseases. To investigate the mechanism by which these repeats may be toxic, we transfected primary rat cerebellar granule neurons with polyglutamine-green fluorescent protein (GFP) fusion constructs containing 19 (Q19-GFP), 35 (Q35-GFP), 56 (Q56-GFP), or 80 (Q80-GFP) glutamine residues. All constructs, except Q19-GFP, aggregated within the nuclei of transfected cells in a length- and time-dependent manner. Although Q35-GFP expression led to the development of several small aggregates per cell, these aggregates were cleared or degraded, and the cells remained viable. In contrast, Q80-GFP expression resulted in one or two large aggregates and induced cell death. Caspase activation was observed after Q80-GFP aggregation, but inhibition of caspases with Boc-aspartyl(OMe)-fluoromethylketone (BAF) only served to delay, not prevent, toxicity. In addition, aggregation and toxicity were not affected by other modulators of neuronal cell death such as genetic deletion of the proapoptotic bcl-2 family member bax or addition of the protein synthesis inhibitor cycloheximide. Lastly, nuclear condensation did not occur as part of the toxicity. These data suggest that polyglutamine-GFP expression is toxic to primary neurons but that the death is distinct from classical apoptosis.

Bax deletion further orders the cell death pathway in cerebellar granule cells and suggests a caspase-independent pathway to cell death.

Dissociated cerebellar granule cells maintained in medium containing 25 mM potassium undergo an apoptotic death when switched to medium with 5 mM potassium. Granule cells from mice in which Bax, a proapoptotic Bcl-2 family member, had been deleted, did not undergo apoptosis in 5 mM potassium, yet did undergo an excitotoxic cell death in response to stimulation with 30 or 100 microM NMDA. Within 2 h after switching to 5 mM K+, both wild-type and Bax-deficient granule cells decreased glucose uptake to <20% of control. Protein synthesis also decreased rapidly in both wild-type and Bax-deficient granule cells to 50% of control within 12 h after switching to 5 mM potassium. Both wild-type and Bax -/- neurons increased mRNA levels of c-jun, and caspase 3 (CPP32) and increased phosphorylation of the transactivation domain of c-Jun after K+ deprivation. Wild-type granule cells in 5 mM K+ increased cleavage of DEVD-aminomethylcoumarin (DEVD-AMC), a fluorogenic substrate for caspases 2, 3, and 7; in contrast, Bax-deficient granule cells did not cleave DEVD-AMC. These results place BAX downstream of metabolic changes, changes in mRNA levels, and increased phosphorylation of c-Jun, yet upstream of the activation of caspases and indicate that BAX is required for apoptotic, but not excitotoxic, cell death. In wild-type cells, Boc-Asp-FMK and ZVAD-FMK, general inhibitors of caspases, blocked cleavage of DEVD-AMC and blocked the increase in TdT-mediated dUTP nick end labeling (TUNEL) positivity. However, these inhibitors had only a marginal effect on preventing cell death, suggesting a caspase-independent death pathway downstream of BAX in cerebellar granule cells.

Sulfated glycans stimulate endocytosis of the cellular isoform of the prion protein, PrPC, in cultured cells.

There is currently no effective therapy for human prion diseases. However, several polyanionic glycans, including pentosan sulfate and dextran sulfate, prolong the incubation time of scrapie in rodents, and inhibit the production of the scrapie isoform of the prion protein (PrPSc), the major component of infectious prions, in cultured neuroblastoma cells. We report here that pentosan sulfate and related compounds rapidly and dramatically reduce the amount of PrPC, the non-infectious precursor of PrPSc, present on the cell surface. This effect results primarily from the ability of these agents to stimulate endocytosis of PrPC, thereby causing a redistribution of the protein from the plasma membrane to the cell interior. Pentosan sulfate also causes a change in the ultrastructural localization of PrPC, such that a portion of the protein molecules are shifted into late endosomes and/or lysosomes. In addition, we demonstrate, using PrP-containing bacterial fusion proteins, that cultured cells express saturable and specific surface binding sites for PrP, many of which are glycosaminoglycan molecules. Our results raise the possibility that sulfated glycans inhibit prion production by altering the cellular localization of PrPC precursor, and they indicate that endogenous proteoglycans are likely to play an important role in the cellular metabolism of both PrPC and PrPSc.

The N-terminal domain of a glycolipid-anchored prion protein is essential for its endocytosis via clathrin-coated pits.

The cellular prion protein (PrPC) is a glycolipid-anchored protein that is involved in the pathogenesis of fatal spongiform encephalopathies. We have shown previously that, in contrast to several other glycolipid-anchored proteins, chPrP, the chicken homologue of mammalian PrPC, is endocytosed via clathrin-coated pits in cultured neuroblastoma cells, as well as in embryonic neurons and glia (Shyng, S.-L., Heuser, J. E., and Harris, D. A. (1994) J. Cell Biol. 125, 1239-1250). In this study, we have determined that the N-terminal half of the chPrP polypeptide chain is essential for its endocytosis. Deletions within this region reduce the amount of chPrP internalized, as measured by surface iodination or biotinylation, and decrease its concentration in clathrin-coated pits, as determined by quantitative electron microscopic immunogold labeling. Mouse PrP, as well as two mouse PrP/chPrP chimeras, are internalized as efficiently as chPrP, suggesting that conserved features of secondary and tertiary structure are involved in interaction with the endocytic machinery. Our results indicate that the ectodomain of a protein can contain endocytic targeting information, and they strongly support a model in which the polypeptide chain of PrPC binds to the extracellular domain of a transmembrane protein that contains a coated pit localization signal in its cytoplasmic tail.