Skeletal muscle in healthy humans exhibits a day-night rhythm in lipid metabolism

Abstract

Objective: Human energy metabolism is under the regulation of the molecular circadian clock; we recently reported that mitochondrial respiration displays a day-night rhythm under study conditions that are similar to real life. Mitochondria are interconnected with lipid droplets, which are of importance in fuel utilization and play a role in muscle insulin sensitivity. Here, we investigated if skeletal muscle lipid content and composition also display day-night rhythmicity in healthy, lean volunteers.

Methods: Skeletal muscle biopsies were obtained from 12 healthy lean male volunteers every 5 h over a 24 h period. Volunteers were provided with standardized meals, and biopsies were taken 4.5 h after each last meal. Lipid droplet size and number were investigated by confocal microscopy. Additionally, the muscle lipidome was assessed using UPLC/HRMS-based semi-targeted lipidomics.

Results: Confocal microscopy revealed diurnal differences in intramyocellular lipid content (P < 0.05) and lipid droplet size in oxidative type 1 muscle fibers (P < 0.01). Lipidomics analysis revealed that 13% of all detected lipids displayed significant day-night rhythmicity. The most rhythmic lipid species were glycerophospholipids and diacylglycerols (DAG), with the latter being the largest fraction (>50% of all rhythmic species). DAG levels showed a day-night pattern with a trough at 1 PM and a peak at 4 AM.

Conclusions: Using two distinct methods, our findings show that myocellular lipid content and whole muscle lipid composition vary across the day-night cycle under normal living conditions. In particular, day-night rhythmicity was present in over half of the DAG lipid species. Future studies are needed to investigate whether rhythmicity in DAG is functionally related to insulin sensitivity and how this might be altered in prediabetes.

Keywords: Circadian clock; Human skeletal muscle; Lipid metabolism; Lipidomics.

Figure 1

Lipid droplet morphology shows diurnal variation. (A) Workflow for lipid droplet analysis using lipid staining and confocal microscopy. (B) Predominantly oxidative type I fiber lipid droplet size and (D) number show day-night rhythmicity. (C) Predominantly glycolytic type II fibers show diurnal variations in lipid droplet size and (E) number that are not significant. Lipid droplets (green) and cell membranes (blue) were stained and quantified. (F) Representative images of type I fiber lipid droplet size of one subject for the different time points are depicted. Grey area represents sleeping periods (11 PM–7 AM). ∗P ≤ 0.05 for the effect of time.

Figure 2

Semi-targeted lipidomic analysis reveals differences in lipid species abundance over 24 h. (A) The workflow of lipidomic analysis using UPLC-HRMS and bioinformatic pipeline. (B) Overview of lipid clusters that are divided into separate lipid classes according to their chemical properties. Heatmaps show z-scores of all detected lipids at 8 AM, 1 PM, 6 PM, 11 PM, and 4 AM. Each time point is the average of all 12 subjects. Lipid species are clustered into (C) sterol lipids and sphingolipids, (D) diradylglycerols and triradylglycerols, and (E) glycerophospholipids.

Figure 3

(A) Rhythmic regulation of the skeletal muscle lipidome. Composition of the lipidome (rhythmic and nonrhythmic lipids) according to the five main lipid clusters. (B) Overview of rhythmic lipids and distribution in the main lipid clusters. (C) Bar graph showing the percentage of rhythmic lipids per lipid cluster. More than half of all diradylglycerols are classified as being rhythmic. (D–H) Pattern per lipid cluster shows the average ± SEM per timepoint of all rhythmic lipid species within the cluster. (D) Diradylglycerols, (E) triradylglycerols, (F) glycerophospholipids, (G) sphingolipids, and (H) sterols. (I) Overview of peak times of all lipid clusters. Grey area represents sleeping periods (11 PM–7 AM).

Figure 4

Analysis of individual lipid classes. Rhythmic lipid species in glycerophospholipids in which saturation or chain length has no influence on the pattern such as (A) phosphatidylcholine (PC) and its alkyl-containing counterpart (PC[O]) (B) as well as phosphatidylinositols (PI). Rhythmic lipid species show differences based on saturation or carbon chain length in the subclasses (C) Bis(monoacylglycero)phosphate (BMP), (D) Alkylphosphatidylethanolamine (PE[O]), (E) Phosphatidylethanolamine (PE), and (F). phosphatidylserines (PS). In (G) hexosylceramides (HexCer) and (H) ceramides (CER) showed an opposing rhythmic profile. Black lines depict lipid species that follow a similar pattern within the lipid class. Blue lines depict saturated lipids and red lines depict unsaturated lipids. Grey area represents sleeping periods (11 PM–7 AM).