Comparative Study of the Metabolic Pools of Sphingomyelin and Phosphatidylcholine Sensitive to Tumor Necrosis Factor

Abstract

The metabolism and localization of the pools of sphingomyelin and phosphatidylcholine (PtdCho) which are hydrolyzed upon activation of the sphingomyelin signal transduction pathway were studied in human skin fibroblasts treated with tumor necrosis factor a (TNF-a). In a first series of experiments, cellular phospholipids were labeled with [3H]choline under conditions that inhibit the vesicular traffic to the plasma membrane. Thus, in human fibroblasts metabolically labeled in the presence of brefeldin A, monensin or at 20°C, the arrival of newly synthesized sphingomyelin to the cell surface was prevented, supporting previous conclusions for a vesicular mechanism of sphingomyelin transport to the plasma membrane. Under these conditions, TNF-a induced the hydrolysis of PtdCho but did not promote the hydrolysis of 3H-labeled sphingomyelin, suggesting that the sphingomyelin signaling pool resides in a compartment distal to the Golgi apparatus, and possibly in the plasma membrane. TNF was also unable to trigger the breakdown of a radioactive sphingomyelin, [cerumide-'Hlsphingomyelin, exogenously added to the cells to label the exoplasmic side of the cell surface. However, TNF caused PtdCho and sphingomyelin degradation in fibroblasts that had been treated with bacterial sphingomyelinase to degrade the sphingomyelin pool of the external leaflet of the plasma membrane. A similar result was obtained at 4"C, i.e. under conditions which inhibit endocytosis, thereby excluding the endosomes as a potential site for TNF-induced sphingomyelin hydrolysis. Altogether, these results strongly argue for a localization of the sphingomyelin signaling pool at the inner leaflet of the plasma membrane, but neither in the endolysosoma1 nor the Golgi compartments. In addition, when ['Hlcholine-labeled fibroblasts were treated under non-lytic conditions with bacterial phospholipase C to degrade the external pool of PtdCho, TNF was still able to stimulate the hydrolysis of PtdCho. This demonstrates that the pool of PtdCho involved in TNF-a signaling (and which is hydrolyzed concurrently with sphingomyelin to generate diacylglycerol), is not located in the outer leaflet of the plasma membrane.

Caspases are the main executioners of Fas-mediated apoptosis, irrespective of the ceramide signalling pathway

“…Treatment of Jurkat and U937 cells with a combination of monensin and EDTA, in the absence of TNF and anti-Fas, caused a greater increase in the ceramide to sphingomyelin ratio (47 and 66%, respectively). This fact was probably due to the intracellular accumulation of ceramide by monensin blockade of its transport to the plasma membrane (Kallen et al, 1993;Andrieu et al, 1996). This increase in the cell ceramide content was not associated, however, with any observable cytotoxicity during the experimental period, in spite of being greater than that induced by anti-Fas or TNF (Figure 2).…”

“…The effect of several drugs, which interfere with ceramide metabolism, on TNF and Fasmediated apoptosis was also evaluated. These were FB 1 (25 ± 50 mM), an inhibitor of ceramide synthase (Jaffre  zou et al, 1996), the lysosomotropic agents chloroquine (10 mM), ammonium chloride (8 mM) and monensin (1 ± 10 mg/ml), all of which inhibit acid sphingomyelinase activity (Wiegmann et al, 1994;Tepper et al, 1995;Andrieu et al, 1996), and EDTA (1 mM), a Ca 2+ -and Mg 2+ -chelator which inhibits the activity of neutral, Mg 2+ -dependent, sphingomyelinase (Futerman et al, 1990). Chloroquine was dissolved in RPMI medium, EDTA and FB 1 in PBS and monensin in ethanol.…”

“…To evaluate the contribution to apoptosis of sphingomyelinase-generated ceramide, as a first approach, U937 and Jurkat cells were treated with cytotoxic anti-Fas antibodies or TNF in the presence or absence of inhibitors of neutral (EDTA) (Wiegmann et al, 1994;Tepper et al, 1995) and acidic (ammonium chloride, chloroquine and monensin) (Wiegmann et al, 1994;Andrieu et al, 1996) sphingomyelinases. Since incubations with these compounds (at the concentrations reported to be inhibitory) for the times required for physiological cell lysis (8 ± 12 h for anti-Fas, 24 ± 48 h for TNF), produced a high toxicity, tests were performed in the presence of cycloheximide to accelerate the onset of cell death (Schulze-Osthoff et al, 1994) (Figure 1).…”

Relationships of apoptotic signaling mediated by ceramide and TNF-α in U937 cells

“…4,46,47 Interestingly, the location of the sphingomyelin pool hydrolyzed by TNF-a is controversial. Several investigators have proposed that the TNF-asensitive pool resides in the external leaflet of the plasma membrane.…”

“…48 Others proposed that the sphingomyelin pool hydrolyzed by TNF-a is localized in the endolysosomal compartments 4 or the inner leaflet of the plasma membrane. 47,49 One could argue that exogenous ceramide kills cells with different kinetics, because of its differential accumulation in subcellular compartments that are different from those generated by TNF-a. Studies with synthetic fluorescent ceramide analogs have shown ceramide to accumulate in the Golgi apparatus and sequentially in other membrane compartments 39 suggesting that cellular membranes are saturated quickly in the presence of exogenous ceramide.…”