Serum dihydroceramides correlate with insulin sensitivity in humans and decrease insulin sensitivity in vitro

Abstract

Serum ceramides, especially C16:0 and C18:0 species, are linked to CVD risk and insulin resistance, but details of this association are not well understood. We performed this study to quantify a broad range of serum sphingolipids in individuals spanning the physiologic range of insulin sensitivity and to determine if dihydroceramides cause insulin resistance in vitro. As expected, we found that serum triglycerides were significantly greater in individuals with obesity and T2D compared with athletes and lean individuals. Serum ceramides were not significantly different within groups but, using all ceramide data relative to insulin sensitivity as a continuous variable, we observed significant inverse relationships between C18:0, C20:0, and C22:0 species and insulin sensitivity. Interestingly, we found that total serum dihydroceramides and individual species were significantly greater in individuals with obesity and T2D compared with athletes and lean individuals, with C18:0 species showing the strongest inverse relationship to insulin sensitivity. Finally, we administered a physiological mix of dihydroceramides to primary myotubes and found decreased insulin sensitivity in vitro without changing the overall intracellular sphingolipid content, suggesting a direct effect on insulin resistance. These data extend what is known regarding serum sphingolipids and insulin resistance and show the importance of serum dihydroceramides to predict and promote insulin resistance in humans.

Keywords: CVD; T2D; circulating ceramides; insulin resistance; lipidomics; myotube; obesity; serum; sphingolipids.

Figure 1

Serum triglyceride content in individuals spanning different ranges of insulin sensitivities. Serum triglycerides (triacylglycerol [TAG]) in endurance trained athletes, lean sedentary controls, and individuals with obesity without and with T2D. Values are means ± SEM and expressed as ratio between the analyte and the internal standard areas in 1 ml of serum. Values are means ± SEM. ¥ = significantly different than lean P < 0.05, # = significantly different than athletes P < 0.05.

Figure 2

Serum ceramides and relationships to insulin sensitivity. Serum ceramide content (A) and relationship of C18:0 (B), C20:0 (C), and C22:0 (D) ceramide species to insulin sensitivity in endurance trained athletes, lean sedentary controls, and individuals with obesity without and with T2D. Values are means ± SEM.

Figure 3

Serum dihydroceramides and relationships to insulin sensitivity. Serum dihydroceramide content (A) and relationship of total dihydroceramide content to insulin sensitivity (B) in endurance trained athletes, lean sedentary controls, and individuals with obesity without and with T2D. Relationships of individual dihydroceramide species to insulin sensitivity are shown for C18:0 (C), C20:0 (D), C22:0 (E), C24:0 (F), and C24:1 (G). Values are means ± SEM. ¥ = significantly different than lean P < 0.05, # = significantly different than athletes P < 0.05.

Figure 4

Effect of dihydroceramide-containing liposome administration on insulin sensitivity and cellular sphingolipid accumulation in human primary myotubes in vitro. Cells were treated for 3.5 h with POPC liposomes containing vehicle control (control) or a mixture of C18:0, C24:0, and C24:1 dihydroceramides (DHCer mix) (A) or individual dihydroceramides at the same concentration as the DHCer mix treatment (B). Effect of dihydroceramide administration on cellular sphingolipid content (C). Values are means ± SEM. § = significantly different than control, P < 0.05.

Figure 5

Effect of dihydroceramide administration on basal and insulin-stimulated signaling and inflammation in cultured primary human myotubes. Basal versus insulin-stimulated insulin signaling (A), relative changes in phosphorylation of downstream effectors of insulin signaling (B), inflammatory response (C), and representative blots for insulin signaling (D), and inflammatory response (E) in human skeletal muscle cells treated with POPC control liposomes (control), or with POPC liposomes containing a mixture of C18:0, C24:0, and C24:1 dihydroceramides (DHCer mix). Values are means ± SEM. ¥ = significantly different than basal, P < 0.05.