Regulation of SNARE fusion machinery by fatty acids.

Vesicle fusion is a ubiquitous biological process involved in membrane trafficking and a variety of specialised events such as exocytosis and neurite outgrowth. The energy to drive biological membrane fusion is provided by fusion proteins called SNAREs. Indeed, SNARE proteins play critical roles in neuronal development as well as neurotransmitter and hormone release. SNARE proteins form a very tight alpha-helical bundle that can pull two membranes together, thereby initiating fusion. Whereas a great deal of attention has been paid to partner proteins that can affect SNARE function, recent genetic and biochemical evidence suggests that local lipid environment may be as important in SNARE regulation. Direct lipid modification of SNARE fusion proteins and their regulation by fatty acids following phospholipase action will be discussed here in detail. Our analysis highlights the fact that lipids are not a passive platform in vesicle fusion but intimately regulate SNARE function.

SNAP25 is a pre-synaptic target for the depressant action of reactive oxygen species on transmitter release.

Reactive oxygen species (ROS) participate in various physiological and pathological processes in the nervous system, but the specific pathways that mediate ROS signalling remain largely unknown. Using electrophysiological techniques and biochemical evaluation of isolated fusion proteins, we explored the sensitivity to standard oxidative stress of the entire synapse, the pre-synaptic machinery and essential fusion proteins underlying transmitter exocytosis. Oxidative stress induced by H(2)O(2) plus Fe(2+) inhibited both evoked and spontaneous quantal release from frog or mouse motor nerve endings, while it left post-synaptic sensitivity unchanged. The depressant effect of H(2)O(2) on acetylcholine release was pertussis toxin-insensitive, ruling out G-protein cascades. Experiments with ionomycin, a Ca(2+) ionophore, revealed that ROS directly impaired the function of releasing machinery. In line with this, SNAP25, one of three essential fusion proteins, showed a selectively high sensitivity to the oxidative signals. Several ROS scavengers enhanced evoked synaptic transmission, consistent with tonic inhibition by endogenous ROS. Our data suggest that ROS-induced impairment of releasing machinery is mediated by SNAP25, which appears to be a pre-synaptic ROS sensor. This mechanism of ROS signalling could have widespread implications in the nervous system and might contribute to the pathogenesis of neurodegenerative diseases.

Parkin modulates gene expression in control and ceramide-treated PC12 cells.

Mutations in the parkin gene cause autosomal-recessive early-onset parkinsonism as a result of the degeneration of mesencephalic dopaminergic neurons. In cell culture models, parkin expression has been shown to protect against cell death mediated by the sphingolipid ceramide. To determine whether the antiapoptotic effect of parkin involves changes in gene expression, we used Affymetrix oligonucleotide microarrays to analyse gene expression in stably transfected PC12 cells which conditionally overexpress parkin, that were treated or not with C2-ceramide. Overexpression of parkin and ceramide treatment both modulated gene expression. A number of the genes upregulated in the presence of ceramide, and modulated by parkin, were associated with apoptosis or cellular stress reactions. We validated the upregulation of four such genes (CHK, EIF4EBP1, GADD45A and PTPN-5) by real-time PCR after 3, 6, 9 and 12 h of ceramide treatment in cells that overexpressed parkin or not. All were upregulated 2 to 11-fold, 3 and 6 h after application of ceramide. Parkin overexpression reduced the upregulation of EIF4EBP1, GADD45A and PTPN-5, but only at 6 h. These results suggest that, in this assay, the cytoprotective effect of parkin might result not only from its E3-ligase activity, but also from direct or indirect modulation of gene expression in a time-dependent manner.

The mitochondrial complex I inhibitor annonacin is toxic to mesencephalic dopaminergic neurons by impairment of energy metabolism.

The death of dopaminergic neurons induced by systemic administration of mitochondrial respiratory chain complex I inhibitors such as 1-methyl-4-phenylpyridinium (MPP(+); given as the prodrug 1-methyl-1,2,3,6-tetrahydropyridine) or the pesticide rotenone have raised the question as to whether this family of compounds are the cause of some forms of Parkinsonism. We have examined the neurotoxic potential of another complex I inhibitor, annonacin, the major acetogenin of Annona muricata (soursop), a tropical plant suspected to be the cause of an atypical form of Parkinson disease in the French West Indies (Guadeloupe). When added to mesencephalic cultures for 24 h, annonacin was much more potent than MPP(+) (effective concentration [EC(50)]=0.018 versus 1.9 microM) and as effective as rotenone (EC(50)=0.034 microM) in killing dopaminergic neurons. The uptake of [(3)H]-dopamine used as an index of dopaminergic cell function was similarly reduced. Toxic effects were seen at lower concentrations when the incubation time was extended by several days whereas withdrawal of the toxin after a short-term exposure (<6 h) arrested cell demise. Unlike MPP(+) but similar to rotenone, the acetogenin also reduced the survival of non-dopaminergic neurons. Neuronal cell death was not excitotoxic and occurred independently of free radical production. Raising the concentrations of either glucose or mannose in the presence of annonacin restored to a large extent intracellular ATP synthesis and prevented neuronal cell demise. Deoxyglucose reversed the effects of both glucose and mannose. Other hexoses such as galactose and fructose were not protective. Attempts to restore oxidative phosphorylation with lactate or pyruvate failed to provide protection to dopaminergic neurons whereas idoacetate, an inhibitor of glycolysis, inhibited the survival promoting effects of glucose and mannose indicating that these two hexoses acted independently of mitochondria by stimulating glycolysis. In conclusion, our study demonstrates that annonacin promotes dopaminergic neuronal death by impairment of energy production. It also underlines the need to address its possible role in the etiology of some atypical forms of Parkinsonism in Guadeloupe.

Noradrenaline provides long-term protection to dopaminergic neurons by reducing oxidative stress.

To better understand the neurotrophic function of the neurotransmitter noradrenaline, we have developed a model of mesencephalic cultures in which we find low concentrations (0.3-10 microM) of noradrenaline to be remarkably effective in promoting long-term survival and function of dopaminergic neurons. This protective action reproduced the effect of caspase inhibition. It was atypical in that it occurred independently of adrenoceptor activation and was mimicked by some antioxidants, redox metal chelators and the hydroxyl radical detoxifying enzyme catalase. Interestingly, intracellular reactive oxygen species (ROS) were drastically reduced by treatment with noradrenaline, indicating that the neurotransmitter itself acted as an antioxidant. Prevention of oxidative stress was, however, independent of the glutathione antioxidant defense system. Chemical analogues of noradrenaline bearing two free hydroxyl groups in the ortho position of the aromatic ring (o-catechols), as well as o-catechol itself, mimicked the survival promoting effects of the neurotransmitter, suggesting that this diphenolic structure was critical for both neuroprotection and reduction of ROS production. Paradoxically, the autoxidation of noradrenaline and the ensuing production of quinone metabolites may be required for both effects, as the neurotransmitter was spontaneously and rapidly degraded over time in the culture medium. These results support the concept that central noradrenergic mechanisms have a neuroprotective role, perhaps in part by reducing oxidative stress.

Mitochondrial free calcium levels (Rhod-2 fluorescence) and ultrastructural alterations in neuronally differentiated PC12 cells during ceramide-dependent cell death.

Mitochondrial free calcium levels measured by Rhod-2 fluorescence and ultrastructure were examined during cell death in nerve growth factor (NGF)-differentiated PC12 cells that were 1) exposed to C2-ceramide, 2) deprived of serum to induce endogenous ceramide production, or 3) treated with calcium ionophore A23187. Rhod-2 fluorescence in mitochondria and also in the nucleolus increased to a maximum within 3 hours after C2-ceramide treatment or serum withdrawal. In A23187-treated cells, Rhod-2 fluorescence remained at baseline levels. In all three models, enlargement of the endoplasmic reticulum was the first ultrastructural alteration, followed by mitochondrial shrinkage in ionophore-treated cells, but by mitochondrial swelling in the ceramide-dependent models, in which rupture of the outer mitochondrial membrane and unfolding of the inner membrane were frequently seen. Dihydro-C2-ceramide, which did not cause cell death, had no effect on cellular ultrastructure. NGF, which inhibits ceramide-dependent cell death, prevented the effects of serum deprivation on mitochondrial ultrastructure but not on endoplasmic reticulum morphology or Rhod-2 fluorescence. Nuclear shrinkage with loss of nuclear membrane integrity, characterized by nuclear pores, free or surrounded by electron-dense filaments, was a late event in ceramide-dependent cell death. Chromatin condensation and other morphological features associated with apoptosis were seen in only a few atypical cells. Ceramide-mediated cell death, therefore, did not involve classical apoptosis but was mediated by a reproducible series of events beginning in the endoplasmic reticulum, followed by the mitochondria, and then the nucleus. NGF-dependent cell death inhibition intervenes at the mitochondrial level, not by blocking the increase in Rhod-2 fluorescence but by preventing the ultrastructural changes that follow.