The mechanism by which opiates affect fetal development is unknown, but one potential target is the programmed cell death (apoptosis) pathway of neurons. Apoptosis was induced in both primary neuronal cultures from embryonic day 7 cerebral hemispheres of chick brain (E7CH) and the F‐11κ7 cell line (an immortalized mouse neuroblastoma × dorsal root ganglion hybrid stably transfected to overexpress κ‐opioid receptors) by either staurosporine or the phosphatidylinositol 3‐kinase inhibitors wortmannin and LY294002. Cells pretreated with either the μ‐specific opioid agonist morphiceptin (E7CH) or the κ‐specific opioid agonist U69,593 (F‐11κ7) for 24 h showed increased apoptosis in response to staurosporine or wortmannin when compared with nonpretreated cells. The effects of morphiceptin and U69,593 were time‐ and dose‐dependent and antagonist‐reversible, suggesting that they were receptor‐mediated. Neither morphiceptin nor U69,593 by themselves had any measurable effect on cell viability or DNA fragmentation, and coaddition of opiates at the same time as staurosporine, wortmannin, or LY294002 did not enhance apoptosis. Time course studies indicated a maximal opioid effect at a time (16–24 h) when inhibition of adenylate cyclase had been maximal for many hours. Addition of dibutyryl cyclic AMP either before or at the time of opioid addition protected against apoptosis and reduced fragmentation to levels seen for staurosporine plus dibutyryl cyclic AMP alone. The specificity for cyclic AMP was confirmed by showing protection with the specific agonist Sp‐adenosine 3′,5′‐cyclic monophosphothioate and increased killing with the antagonist Rp‐adenosine 3′,5′‐cyclic monophosphothioate. We conclude that the opioid enhancement of apoptosis is based on the inhibition of adenylate cyclase and that the effect is time‐dependent.

An immortalized dorsal root ganglion cell line F-11 exhibits many properties of spinal cord neurons and undergoes apoptosis in response to growth factor withdrawal and the exogenous addition of inhibitors of phosphatidylinositol-3-kinase (PI3K). To elucidate the mechanism of apoptosis we generated F-11 clones which overexpressed either the p110 subunit of PI3K, a constitutively active form of protein kinase B/Akt (Myristoylated Akt), or a dominant-negative form (c-Akt). The first two constructs were protective against apoptosis induced by PI3K inhibitors such as wortmannin and LY294002. Caspase-3 (CPP32) levels peaked at 4 hr to 6 hr in response to pro-apoptotic drugs, and this increase was attenuated by 50% in F-11 with constitutively active Akt. The Akt protection was confirmed by DNA fragmentation studies. Both neo-transfected and the c-Akt dominant-negative transfected F-11 cells showed increased ceramide formation (twofold) in response to staurosporine, wortmannin, or LY294002; whereas cells with a constitutively active Akt (Myr-Akt) showed no increase in ceramide when treated with staurosporine, wortmannin, or LY294002. Ceramide was a more potent activator of CPP32 and an inducer of apoptosis when added as the native form (hydroxy- or nonhydroxy-), rather than the more water-soluble C(2)-ceramide. Overexpression of PI3K (p110) and Akt protected cells against ceramide-induced apoptosis, suggesting that Ceramide action is upstream of Akt in these cells and suggesting that Akt might be a target for inhibition by ceramide. Both staurosporine and C(2)-ceramide activated the Jun kinase (JNK) cascade and C(2)-ceramide increased caspase-3 (CPP32) activity in cells expressing wild-type c-Jun, but not dominant-negative (TAM-67) c-Jun. We suggest that this pathway is also involved in apoptosis, consistent with the idea that ceramide has multiple kinase and kinase-modulating targets in the apoptotic pathway of neurons. J. Neurosci. Sci. 57:884-893, 1999.

Demyelination is a common result of oxidative stress in the nervous system, and we report here that the response of oligodendrocytes to oxidative stress involves the receptor for advanced glycation end products (RAGE). RAGE has not previously been reported in neonatal rat oligodendrocytes (NRO), but, by using primers specific for rat RAGE, we were able to show expression of messenger RNA (mRNA) for RAGE in NRO, and a 55-kDa protein was detected by Western blotting with antibodies to RAGE. Neonatal rat oligodendrocytes stained strongly for RAGE, suggesting membrane localization of RAGE. Addition of low concentrations of hydrogen peroxide (100 μM) initiated 55-kDa RAGE shedding from the cell membrane and the appearance of "soluble" 45-kDa RAGE in the culture medium, followed by restoration of RAGE expression to normal levels. Increasing hydrogen peroxide concentration (>200 μM) resulted in no restoration of RAGE, and the cells underwent apoptosis and necrosis. We further confirmed the observation in a human oligodendroglioma-derived (HOG) cell line. Both the antioxidant N-acetyl-L-cysteine and the broad-spectrum metalloproteases inhibitor TAPI0 were able partially to inhibit shedding of RAGE, suggesting involvement of metalloproteases in cleavage to produce soluble RAGE. The level of 55-kDa RAGE in autopsy brain of patients undergoing neurodegeneration with accompanying inflammation [multiple sclerosis and neuronal ceroid-lipofuscinosis (Batten's disease)] was much lower than that in age-matched controls, suggesting that shedding of RAGE might occur as reactive oxygen species accumulate in brain cells and be part of the process of neurodegeneration. Keywordsoligodendrocytes; reactive oxygen species; hydrogen peroxide; RAGE; shedding proteolysis Oligodendrocytes have been implicated in the pathogenesis of demyelinating diseases such as multiple sclerosis (MS) in which the active agents are reactive oxygen species (ROS) and the products of inflammation (LeVine, 1992;Smith et al., 1999). Evidence of ROS damage has been reported in demyelinating lesions in the brains of MS patients (Langemann et al., 1992;LeVine and Wetzel, 1998) and in the demyelinating experimental allergic encephalitis (EAE) mouse spinal cord (MacMicking et al., 1992). Lipid peroxidation products and ROS have been identified in cerebrospinal fluid (CSF), plasma, and brain from MS patients (Naidoo and Knapp, 1992;LeVine, 1992;Qin et al., 2007), and this, together with evidence for peroxynitrite modification of proteins (Liu et al., 2001) NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author Manuscript 1998), strongly supports the view that oxidative damage is important in demyelination. Protein carbonyl content, mainly glutamic semialdehyde (from arginine and proline) and aminoadipic semialde-hyde (from lysine oxidation), is a measure of ROS-mediated protein oxidation (Berlett and Stadtman, 1997) and is increased in MS brain (Bizzozero et al., 2005).RAGE is a 45−55-kDa member of the immunoglobulin superfamily (Rong et al., 2004a,b;...

Cell differentiation and myelination involve a fine balance between stasis and programmed cell death, yet the genes that regulate this have not been clearly defined. We therefore studied two key gene products involved in oligodendrocyte plasma membrane lipid metabolism and their antagonistic role in ceramide-mediated cell death signaling. Overexpression of palmitoyl:protein thioesterase (PPT1; verified by Western blot of the V5-tagged protein and increased enzyme activity) resulted in decreased ceramide in the detergent-resistant microdomain (DRM, or raft) relative to cholesterol and sphingomyelin (SM). This PPT1 overexpression also resulted in protection against cell death induced by either staurosporine or C(2)-ceramide. In contrast, overexpression of neutral sphingomyelinase 2 (NSMase2; verified by Western blot of the FLAG-tagged protein and increased enzyme activity) resulted in increased membrane NSMase and increased ceramide in rafts relative to cholesterol and SM. The difference in SM and ceramide turnover was quantitated by [(3)H]palmitate pulse-chase labeling. Furthermore, when NBD-SM was added to cells, it was hydrolyzed by NSMase-transfected cells at more than twofold the rate in untransfected cells. NSMase2 overexpression enhanced cell death induced by staurosporine or C(2)-ceramide, in contrast to the protective effect of PPT1 overexpression. The presence of a fraction of both PPT1 and NSMase2 in rafts together with their substrates (palmitoylated proteins and SM, respectively) suggests a mechanism for dynamic palmitoylation/depalmitoylation of certain proteins in controlling cell death via NSMase activation.

Chronic exposure of embryonic brain to opioids leads to microcephaly and developmental abnormalities. An immortalized mouse neuroblastoma X dorsal root ganglion hybrid cell line stably transfected to overexpress K-opioid receptors (F-1 lK 7) showed complete loss of K-receptor binding to [3H]U69,593after exposure to the K-agonist U69,593 for 24 h. U69,593 had no measurable effect on cell viability as determined by either cell viability or DNA fragmentation assays. However, when cell death (apoptosis) was induced by either staurosporine or the phosphatidylinositol 3-kinase inhibitors wortmannin and LY294002, cells pretreated with U69,593 for 24 h showed increased apoptosis compared with untreated cells. Thus, staurosporine (50 nM), wortmannin (4 1sM), and LY294002 (30 ‚uM) treatment for 24 h induced a 50% loss of cell viability and DNA fragmentation in 24 h. U69,593 pretreatment produced the same killing at lower concentrations, namely, 20 nM staurosporine, 2 p.M wortmannin, and 14 ‚uM LY294002, respectively. The effects of U69,593 were time-, dose-, and naloxone-reversible, suggesting that they are receptor-mediated. However, coaddition of U69,593 at the same time as staurosporine, wortmannin, or LY294002 did not enhance apoptosis. All three drugs that induced apoptosis were found to increase the level of ceramide, and pretreatment with U69,593 further increased the rate of formation of ceramide, a lipid that induces apoptosis in cells. We propose that chronic exposure to K-receptor agonists promotes increased vulnerability of neurons to apoptosis.