Exosomal lipid composition and the role of ether lipids and phosphoinositides in exosome biology.

Exosomes are a type of extracellular vesicle released from cells after fusion of multivesicular bodies with the plasma membrane. These vesicles are often enriched in cholesterol, SM, glycosphingolipids, and phosphatidylserine. Lipids not only have a structural role in exosomal membranes but also are essential players in exosome formation and release to the extracellular environment. Our knowledge about the importance of lipids in exosome biology is increasing due to recent technological developments in lipidomics and a stronger focus on the biological functions of these molecules. Here, we review the available information about the lipid composition of exosomes. Special attention is given to ether lipids, a relatively unexplored type of lipids involved in membrane trafficking and abundant in some exosomes. Moreover, we discuss how the lipid composition of exosome preparations may provide useful information about their purity. Finally, we discuss the role of phosphoinositides, membrane phospholipids that help to regulate membrane dynamics, in exosome release and how this process may be linked to secretory autophagy. Knowledge about exosome lipid composition is important to understand the biology of these vesicles and to investigate possible medical applications.

Identification of non-invasive miRNAs biomarkers for prostate cancer by deep sequencing analysis of urinary exosomes.

The aim of this study was to identify microRNAs in urinary exosomes that are differently expressed in prostate cancer patients and healthy donors. For this purpose, RNA was extracted from urinary exosomes from 20 prostate cancer patients and 9 healthy males and the microRNAs were analyzed by next generation sequencing. Interestingly, 5 microRNAs - miR-196a-5p, miR-34a-5p, miR-143-3p, miR-501-3p and miR-92a-1-5p - were significantly downregulated in exosomes from prostate cancer patients. Furthermore, RT-qPCR analysis of an independent cohort of 28 prostate cancer patients and 19 healthy males confirmed that miR-196a-5p and miR-501-3p were downregulated in prostate cancer samples. These results suggest that specific microRNAs in urinary exosomes might serve as non-invasive biomarkers for prostate cancer. In particular, miR-196a-5p and miR-501-3p are promising biomarkers that need to be further studied in large patient cohorts.

A large-scale quantitative EM study on activation of olfactory glands shows no effect of cholinergic agents.

Little is known about olfactory glands' regulation despite their presumed importance for normal functioning of the cilia of olfactory neurons. The aim of this study was to establish an assay for olfactory gland activation by using large-scale quantitative electron microscopy (EM). In addition we wanted to test the hypothesis that cholinergic drugs activate the olfactory glands, by using our newly established EM assay. In total, over 70 000 secretory gland vesicles were quantified in over 3000 cells. Olfactory gland cell size (40.8 µm2 ± 2.0 SD), vesicle diameter (812 nm ± 57 SD) and vesicles per cell (21.6 ± 4.2 SD) were also quantified. The vesicle percentage of the cell area varied between 24% and 30%. In a blinded study we found no significant effects of cholinergic agents on parameters of vesicle number or vesicle diameter. Unexpectedly, pilocarpine treatment increased olfactory gland size, probably by inducing cell swelling. In conclusion, we have established a quantitative EM assay for olfactory gland activation and provided new data on basic olfactory gland cell characteristics. By using the EM assay, olfactory glands are shown not to be activated by cholinergic agents, which indicates an alternative regulation pathway or constitutive secretion from olfactory glands.

Regulation of exosome release by glycosphingolipids and flotillins.

Exosomes are released by cells after fusion of multivesicular bodies with the plasma membrane. The molecular mechanism of this process is still unclear. We investigated the role of sphingolipids and flotillins, which constitute a raft-associated family of proteins, in the release of exosomes. Interestingly, our results show that dl-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol, an inhibitor of glucosylceramide synthase, seemed to affect the composition of exosomes released from PC-3 cells. However, the inhibition of ceramide formation from the de novo pathway by fumonisin B1 did not affect exosome secretion. Moreover, in contrast to findings obtained with other cell lines published so far, inhibition of neutral sphingomyelinase 2, an enzyme that catalyzes the formation of ceramide from sphingomyelin, did not inhibit the secretion of exosomes in PC-3 cells. Finally, small interfering RNA-mediated downregulation of flotillin-1 and flotillin-2 did not significantly change the levels of released exosomes as such, but seemed to affect the composition of exosomes. In conclusion, our results reveal the involvement of glycosphingolipids and flotillins in the release of exosomes from PC-3 cells, and indicate that the role of ceramide in exosome formation may be cell-dependent.

Palmitic acid follows a different metabolic pathway than oleic acid in human skeletal muscle cells; lower lipolysis rate despite an increased level of adipose triglyceride lipase.

Development of insulin resistance is positively associated with dietary saturated fatty acids and negatively associated with monounsaturated fatty acids. To clarify aspects of this difference we have compared the metabolism of oleic (OA, monounsaturated) and palmitic acids (PA, saturated) in human myotubes. Human myotubes were treated with 100µM OA or PA and the metabolism of [(14)C]-labeled fatty acid was studied. We observed that PA had a lower lipolysis rate than OA, despite a more than two-fold higher protein level of adipose triglyceride lipase after 24h incubation with PA. PA was less incorporated into triacylglycerol and more incorporated into phospholipids after 24h. Supporting this, incubation with compounds modifying lipolysis and reesterification pathways suggested a less influenced PA than OA metabolism. In addition, PA showed a lower accumulation than OA, though PA was oxidized to a relatively higher extent than OA. Gene set enrichment analysis revealed that 24h of PA treatment upregulated lipogenesis and fatty acid ß-oxidation and downregulated oxidative phosphorylation compared to OA. The differences in lipid accumulation and lipolysis between OA and PA were eliminated in combination with eicosapentaenoic acid (polyunsaturated fatty acid). In conclusion, this study reveals that the two most abundant fatty acids in our diet are partitioned toward different metabolic pathways in muscle cells, and this may be relevant to understand the link between dietary fat and skeletal muscle insulin resistance.

Chronic hyperglycemia reduces substrate oxidation and impairs metabolic switching of human myotubes.

Skeletal muscle of insulin resistant individuals is characterized by lower fasting lipid oxidation and reduced ability to switch between lipid and glucose oxidation. The purpose of the present study was to examine if chronic hyperglycemia would impair metabolic switching of myotubes. Human myotubes were treated with or without chronic hyperglycemia (20mmol/l glucose for 4 days), and metabolism of [(14)C]oleic acid (OA) and [(14)C]glucose was studied. Myotubes exposed to chronic hyperglycemia showed a significantly reduced OA uptake and oxidation to CO(2), whereas acid-soluble metabolites were increased compared to normoglycemic cells (5.5mmol/l glucose). Glucose suppressibility, the ability of acute glucose (5mmol/l) to suppress lipid oxidation, was 50% in normoglycemic cells and reduced to 21% by hyperglycemia. Adaptability, the capacity to increase lipid oxidation with increasing fatty acid availability, was not affected by hyperglycemia. Glucose uptake and oxidation were reduced by about 40% after hyperglycemia, and oxidation of glucose in presence of mitochondrial uncouplers showed that net and maximal oxidative capacities were significantly reduced. Hyperglycemia also abolished insulin-stimulated glucose uptake. Moreover, ATP concentration was reduced by 25% after hyperglycemia. However, none of the measured mitochondrial genes were downregulated nor was mitochondrial DNA content. Microarray and real-time RT-PCR showed that no genes were significantly regulated by chronic hyperglycemia. Addition of chronic lactate reduced both glucose and OA oxidation to the same extent as hyperglycemia. In conclusion, chronic hyperglycemia reduced substrate oxidation in skeletal muscle cells and impaired metabolic switching. The effect is most likely due to an induced mitochondrial dysfunction.

Oxidation of intramyocellular lipids is dependent on mitochondrial function and the availability of extracellular fatty acids.

Obesity and insulin resistance are related to both enlarged intramyocellular triacylglycerol stores and accumulation of lipid intermediates. We investigated how lipid overflow can change the oxidation of intramyocellular lipids (ICL(OX)) and intramyocellular lipid storage (ICL). These experiments were extended by comparing these processes in primary cultured myotubes established from healthy lean and obese type 2 diabetic (T2D) individuals, two extremes in a range of metabolic phenotypes. ICLs were prelabeled for 2 days with 100 microM [(14)C]oleic acid (OA). ICL(OX) was studied using a (14)CO(2) trapping system and measured under various conditions of extracellular OA (5 or 100 microM) and glucose (0.1 or 5.0 mM) and the absence or presence of mitochondrial uncoupling [carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP)]. First, increased extracellular OA availability (5 vs. 100 microM) reduced ICL(OX) by 37%. No differences in total lipolysis were observed between low and high OA availability. Uncoupling with FCCP restored ICL(OX) to basal levels during high OA availability. Mitochondrial mass was positively related to ICL(OX), but only in myotubes from lean individuals. In all, a lower mitochondrial mass and lower ICL(OX) were related to a higher cell-associated OA accumulation. Second, myotubes established from obese T2D individuals showed reduced ICL(OX). ICL(OX) remained lower during uncoupling (P < 0.001), even with comparable mitochondrial mass, suggesting decreased mitochondrial function. Furthermore, the variation in ICL(OX) in vitro was significantly related to the in vivo fasting respiratory quotient of all subjects (P < 0.02). In conclusion, the rate of ICL(OX) is dependent on the availability of extracellular fatty acids and mitochondrial function rather than mitochondrial mass.