Analyses of the effect of monensin on glycosphingolipid metabolism

Cell density-dependent changes of glycosphingolipid biosynthesis in cultured human skin fibroblasts

In this study, the glycosphingolipid biosynthesis was investigated in the sparse and the confluent cell populations of cultured human skin fibroblasts. The human skin fibroblast cell populations were metabolically pulse labeled with (14)C-galactose (48 h). The amounts of (14)C-radioactivity (cpm) incorporated into extracted and purified total cellular glycosphingolipid fractions were counted by beta-scintillation and the individual glycosphingolipid species were separated by high performance thin layer chromatography and visualized by autoradiography. The relative labeling (%) of individual newly synthesized glycosphingolipid species was detected by densitometric scanning of autoradiographic glycosphingolipid patterns. The incorporation of (14)C-label into total glycosphingolipids per cell increased significantly as the cell-density increased, referring to five fold higher rate of glycosphingolipid biosynthesis de novo in cells at confluency vs. sparse populations. The total newly synthesized glycosphingolipid pattern (100%) of sparse cell populations showed a significant predominance of the gangliosides (70%) over the neutral glycosphingolipids (30%), with ganglioside GM2 as the major species followed by monohexosyl-ceramide. Oppositely, the newly synthesized neutral glycosphingolipids (67%) predominated over the gangliosides (33%) in cells at confluency (contact inhibition). Cells reaching confluency were characterized by: (a) a dramatic increase of absolute amount of all newly synthesized neutral glycosphingolipid species, particularly the most abundant monohexosyl-ceramide and trihexosyl-ceramide, but also of the ganglioside GM3; (b) a drastic decrease of absolute amount of newly synthesized ganglioside GM2. The specific shift in newly synthesized glycosphingolipid pattern in cells reaching confluency suggests a down-regulation of biosynthetic pathway primarily at the level of N-acetylgalactosaminyl-transferase. A possible involvement of glycosphingolipids in cell density-dependent regulation of cell growth through establishment of the direct intermolecular intermembrane interactions is discussed.

Decreased synthesis of glycosphingolipids in cells lacking vimentin intermediate filaments

We are studying defects in glycosphingolipid synthesis in cells lacking vimentin intermediate filaments (vimentin-). Sugars can be incorporated into glycolipids whose ceramide is synthesized either de novo (pathway 1) or from sphingoid bases salvaged from hydrolysis of sphingolipids (pathway 2) and into glycolipids recycling from the endosomal pathway through the Golgi (pathway 3). Vimentin- embryonic fibroblasts, obtained from vimentin-knockout mice, incorporate less sugar into glycolipids than vimentin+ fibroblasts. Using two inhibitors of ceramide synthesis, beta-chloroalanine and fumonisin B1, we found the major defect in synthesis to be in pathway 2 and not in de novo synthesis. We used two additional approaches to analyze the functions of pathways 2 and 3. First, we used exogenous glucosylthioceramide ([14C]C8-Glc-S-Cer), a synthetic, nonhydrolyzable glycosphingolipid, as a precursor for synthesis of larger glycolipids. Vimentin- SW13 cells and embryonic fibroblasts glycosylated [14C]C8-Glc-S-Cer less extensively than their vimentin+ counterparts. Second, we used chloroquine to inhibit the hydrolysis of sphingolipids in endosomes and lysosomes. Chloroquine markedly decreased the incorporation of sugars into glycolipids larger than glucosylceramide. The defect in glycolipid synthesis in vimentin- cells probably results from impaired intracellular transport of glycolipids and sphingoid bases between the endosomal/lysosomal pathway and the Golgi apparatus and endoplasmic reticulum. Intermediate filaments may accomplish this function by contributing to the organization of subcellular organelles and/or by binding proteins that participate in transport processes.

Ceramides are transported through the Golgi apparatus in human keratinocytes in vitro

The intercellular lipid sheets of the stratum corneum constitute the epidermal permeability barrier that permits terrestrial life. Although lamellar granules are known to deliver the precursors of the stratum corneum lipids into the intercellular spaces, their site of origin remains unknown. Lamellar granules have characteristics of both secretory granules and lysosomes, which are known to originate from the Golgi apparatus in other cell types. Glucosylceramides, a major component of lamellar granule contents and the precursors of stratum corneum ceramides, have been found to be synthesized primarily in the early compartments of the Golgi apparatus in other cell types. We have investigated the transport and metabolism of a fluorescently labeled ceramide in human keratinocyte cultures using laser-scanning confocal microscopy and lipid analysis. We found that ceramide is metabolized to glucosylceramide and sphingomyelin as it passes through the Golgi apparatus and the metabolites are then delivered to the plasma membrane. Cold temperature, Brefeldin A, and monensin, all known to inhibit transport from the Golgi to the plasma membrane, prevented ceramide metabolites from appearing at the plasma membrane. Because glucosylceramides are one of the most important lipid constituents of lamellar granules, these results support the hypothesis that the Golgi is the origin of lamellar granules.

Epithelial sphingolipid sorting is insensitive to reorganization of the Golgi by nocodazole, but is abolished by monensin in MDCK cells and by brefeldin A in Caco-2 cells

In epithelial MDCK and Caco-2 cells, short-chain analogs of glucosylceramide and sphingomyelin are delivered from the Golgi to the cell surface with different apical/basolateral polarities, which results in an apical enrichment of the glycolipid glucosylceramide over the phospholipid sphingomyelin. Here, we have interfered with the integrity of the Golgi complex in various ways and tested the effects on lipid transport and sorting. Nocodazole, which depolymerizes microtubules, dispersed the Golgi over the cytoplasm of MDCK cells and reduced transport of newly synthesized C6-NBD-(N-6[7-nitro-2,1,3-benzoxadiazol-4-yl]aminocaproyl)-glucosy lceramide and C6-NBD-sphingomyelin to the apical surface by 40%. The lipids were not mistargeted to the basolateral surface and upon removal of nocodazole, apical transport recovered. Nocodazole did not affect the apical enrichment of glucosylceramide over sphingomyelin. The ionophore monensin led to swelling of the Golgi of MDCK cells and inhibited lipid transport to the cell surface by 30–50%. Whereas sphingomyelin transport to both surface domains was equally affected, monensin mainly inhibited apical transport of glucosylceramide. At 10–20 microM of monensin, the two lipids displayed the same polarity of delivery: sorting between the two lipids was abolished. Brefeldin A at 1 microgram/ml, which resulted in disruption of the Golgi in HepG2 cells and completely inhibited protein secretion, had no inhibitory effect on transport of the C6-NBD-lipids to the surface. The same was observed in Caco-2 cells. However, brefeldin A selectively shifted transport of sphingomyelin towards the apical direction which abolished the apical enrichment of glucosylceramide over sphingomyelin. Caco-2 cells were used because in MDCK cells brefeldin A did not change Golgi structure nor lipid transport and sorting. In summary, modification of the Golgi by monensin and brefeldin A, but not nocodazole, interfered with the sorting event by which glucosylceramide is enriched over sphingomyelin in the transport pathway from the Golgi to the apical surface.

Uptake and metabolism of fluorescent ceramide analogs by rat oligodendrocytes in culture

We studied the metabolism of sphingolipids by oligodendrocytes derived from rat spinal cord by providing lipid vesicles with either N-lissamine-rhodaminyl-ceramide (LRh-Cer) or N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)-ceramide (NBD-Cer) to the cells cultured in a chemically-defined medium. With both probes the major fluorescent product turned out to be sphingomyelin (SM). Most of LRh-SM was not cell-associated but recovered from the culture medium, probably due to back-exchange to the lipid vesicles. The accumulation of LRh-SM, both in the cells and in the medium, was inhibited in the presence of monensin or brefeldin A, whereas the production of NBD-SM was much less affected by these Golgi perturbing drugs. With LRh-Cer as substrate, LRh-labelled fatty acid (FA), galactosyl- and sulfogalactosyl-ceramides (GalCer and SGalCer) were also formed. NBD-Cer, however, was metabolized to glucosylceramide (GlcCer) and GalCer but not to SGalCer or NBD-FA. These data demonstrate that chemical modifications of ceramide alter its metabolism in oligodendrocytes and that the metabolites of LRh-Cer reflect the glycolipid composition of myelin more closely than those of NBD-Cer.