Dietary palmitate and oleate differently modulate insulin sensitivity in human skeletal muscle

Abstract

Aims/hypothesis: Energy-dense nutrition generally induces insulin resistance, but dietary composition may differently affect glucose metabolism. This study investigated initial effects of monounsaturated vs saturated lipid meals on basal and insulin-stimulated myocellular glucose metabolism and insulin signalling.

Methods: In a randomised crossover study, 16 lean metabolically healthy volunteers received single meals containing safflower oil (SAF), palm oil (PAL) or vehicle (VCL). Whole-body glucose metabolism was assessed from glucose disposal (R_d) before and during hyperinsulinaemic-euglycaemic clamps with D-[6,6-²H₂]glucose. In serial skeletal muscle biopsies, subcellular lipid metabolites and insulin signalling were measured before and after meals.

Results: SAF and PAL raised plasma oleate, but only PAL significantly increased plasma palmitate concentrations. SAF and PAL increased myocellular diacylglycerol and activated protein kinase C (PKC) isoform θ (p < 0.05) but only PAL activated PKCɛ. Moreover, PAL led to increased myocellular ceramides along with stimulated PKCζ translocation (p < 0.05 vs SAF). During clamp, SAF and PAL both decreased insulin-stimulated R_d (p < 0.05 vs VCL), but non-oxidative glucose disposal was lower after PAL compared with SAF (p < 0.05). Muscle serine¹¹⁰¹-phosphorylation of IRS-1 was increased upon SAF and PAL consumption (p < 0.05), whereas PAL decreased serine⁴⁷³-phosphorylation of Akt more than SAF (p < 0.05).

Conclusions/interpretation: Lipid-induced myocellular insulin resistance is likely more pronounced with palmitate than with oleate and is associated with PKC isoforms activation and inhibitory insulin signalling.

Trial registration: ClinicalTrials.gov .NCT01736202.

Funding: German Federal Ministry of Health, Ministry of Culture and Science of the State North Rhine-Westphalia, German Federal Ministry of Education and Research, European Regional Development Fund, German Research Foundation, German Center for Diabetes Research.

Keywords: Glucose metabolism; Insulin signalling; Lipotoxicity; Monounsaturated fatty acids; Saturated fat; Skeletal muscle.

Fig. 1

Study design. Lean, healthy adults (10 male and 6 female) randomly ingested either one dose of PAL, SAF or VCL (water) at time point 0 min on three occasions during a period of 12 weeks. Starting at −120 min, d-[6,6-²H₂]glucose was infused up to +480 min. Muscle biopsies were taken at time points −60 min, +120 min, +240 min and +420 min. From +360 min to +480 min, a hyperinsulinaemic–euglycaemic clamp test was performed according to the ‘hot’ glucose infusion (hot-GINF) protocol

Fig. 2

Time courses of circulating lipid metabolites in healthy humans. Plasma concentrations of chylomicrons (a), triacylglycerol (b), total NEFA (c), palmitic acid (d), oleic acid (e) and linoleic acid (f) after ingestion of PAL (red), SAF (blue) or VCL (water, grey) at 0 min. Data are shown as means ± SEM; n = 16 (chylomicrons n = 10). *p < 0.05, **p < 0.01 and ***p < 0.001 vs VCL at same time point; ^†p < 0.05 for PAL vs SAF at same time point (ANOVA adjusted for repeated measures with Tukey–Kramer correction for each time point between interventions). P-basal, pre-basal

Fig. 3

Time courses of circulating hormones and metabolites in healthy humans. Concentrations of plasma GIP (a), plasma GLP-1 (b), plasma glucagon (c), plasma insulin (d), blood glucose (e) and plasma glycerol (f) are presented after ingestion of PAL (red), SAF (blue) or VCL (water, grey) at 0 min. Data are shown as means ± SEM; insulin and blood glucose n = 16; GIP, GLP-1, glucagon and glycerol, n = 4. *p < 0.05, **p < 0.01 and ***p < 0.001 vs VCL at same time point; ^†p < 0.05 and ^††p < 0.01 for PAL vs SAF at same time point (ANOVA adjusted for repeated measures with Tukey–Kramer correction for each time point between interventions). P-basal, pre-basal

Fig. 4

Rates of whole-body glucose disposal (R_d) and EGP during basal and clamp periods in healthy humans. (a, b) During the last 30 min of basal period (+330 min to +360 min), rates of glucose metabolism are presented in the context of the ambient plasma insulin concentration: R_d/insulin (a) and EGP × insulin (b). (c–f) During clamp steady-state (+450 min to +480 min), insulin-stimulated R_d (c), rate of GOX (d), rate of NOXGD (e) and EGP suppression (f) are presented after PAL (red), SAF (blue) or VCL (water, grey) ingestion at 0 min. Data are shown as means ± SEM; n = 16. *p < 0.05, **p < 0.01 and ***p < 0.001 vs VCL; ^†p < 0.05 and ^††p < 0.01 for PAL vs SAF (ANOVA adjusted for repeated measures with Tukey–Kramer correction)

Fig. 5

Myocellular lipid metabolites and insulin signalling (DAG–nPKC pathway) in healthy humans. DAG species 18–1:18–1, 16–0:16–0 and 16–0:18–1 in the cell membrane fraction (a), DAG species 18–1:18–1, 16–0:16–0 and 16–0:18–1 in the lipid droplet fraction (b), nPKCε activation (c), nPKCθ activation (d) and IRS-1 levels (e) as well as serine¹¹⁰¹-phosphorylation of IRS-1 relative to IRS-1 (f) during the pre-basal, basal and clamp periods after ingestion of PAL (red), SAF (blue) or VCL (water, grey) at 0 min. Expression signals on immunoblots are expressed in arbitrary units (AU) after normalising against GAPDH for total and cytosolic proteins and against Na⁺/K⁺-ATPase for membrane proteins. Data are shown as means ± SEM; n = 16 at time point −60 min, n = 10 at +120 min, n = 6 at +240 min and +420 min. *p < 0.05 vs VCL at same time point (ANOVA adjusted for repeated measures with Tukey–Kramer correction for each time point between interventions). LD, lipid droplet fraction; P-basal, pre-basal

Fig. 6

Myocellular lipid metabolites and insulin signalling (ceramide–aPKC pathway) in healthy humans. Ceramide species 16:0, 18:0 and 18:1 in the cell membrane fraction (a), ceramide species 16:0, 18:0 and 18:1 in the lipid droplet fraction (b), aPKCζ activation (c), PP2A (d), Akt (e) and serine⁴⁷³-phosphorylation of Akt relative to Akt (f) after ingestion of PAL (red), SAF (blue) or VCL (water, grey) at 0 min. Expression signals on immunoblots are expressed in arbitrary units (AU) after normalising against GAPDH for total and cytosolic proteins and against Na⁺/K⁺-ATPase for membrane proteins. Data are shown as means ± SEM; n = 16 at time point −60 min, n = 10 at +120 min, n = 6 at +240 min and +420 min. *p < 0.05 vs VCL at same time point; ^†p < 0.05 for PAL vs SAF at same time point (ANOVA adjusted for repeated measures with Tukey–Kramer correction for each time point between interventions). P-basal; pre-basal