Treatment of human skeletal muscle cells with inhibitors of diacylglycerol acyltransferases 1 and 2 to explore isozyme-specific roles on lipid metabolism

Abstract

Diacylglycerol acyltransferases (DGAT) 1 and 2 catalyse the final step in triacylglycerol (TAG) synthesis, the esterification of fatty acyl-CoA to diacylglycerol. Despite catalysing the same reaction and being present in the same cell types, they exhibit different functions on lipid metabolism in various tissues. Yet, their roles in skeletal muscle remain poorly defined. In this study, we investigated how selective inhibitors of DGAT1 and DGAT2 affected lipid metabolism in human primary skeletal muscle cells. The results showed that DGAT1 was dominant in human skeletal muscle cells utilizing fatty acids (FAs) derived from various sources, both exogenously supplied FA, de novo synthesised FA, or FA derived from lipolysis, to generate TAG, as well as being involved in de novo synthesis of TAG. On the other hand, DGAT2 seemed to be specialised for de novo synthesis of TAG from glycerol-3-posphate only. Interestingly, DGAT activities were also important for regulating FA oxidation, indicating a key role in balancing FAs between storage in TAG and efficient utilization through oxidation. Finally, we observed that inhibition of DGAT enzymes could potentially alter glucose-FA interactions in skeletal muscle. In summary, treatment with DGAT1 or DGAT2 specific inhibitors resulted in different responses on lipid metabolism in human myotubes, indicating that the two enzymes play distinct roles in TAG metabolism in skeletal muscle.

Figure 1

Effect of DGAT inhibitors on incorporation of oleic acid into various lipid classes. Human myotubes were grown and differentiated on 12- or 24-well tissue culture plates for 7 days. On day 7 of differentiation myotubes were incubated with 100 µM [¹⁴C]oleic acid (0.5 µCi/ml) for 4 h in presence or absence of D1i (1 µM) and/or D2i (10 µM). Cellular lipids were extracted and separated by thin layer chromatography (TLC), and radioactivity was measured by liquid scintillation. The sum of all radioactivity on the TLC plate is presented as total lipids. (A–F) Lipid distribution of [¹⁴C]oleic acid after treatment with D1i, D2i and combination of D1i and D2i. Results are presented as mean ± SEM as nmol/mg protein (A,C,E) or as % of total lipids in the cell (B,D,F). n = 6 (A,B), n = 8 (C,D) and n = 5 (E,F) individual experiments with 3 separate culture wells for each condition. *p < 0.05 vs. control, paired t-test. CE, cholesteryl ester; D1i, DGAT1 inhibitor; D2i, DGAT2 inhibitor; DAG, diacylglycerol; FFA, free fatty acid; PL, phospholipid; TAG, triacylglycerol.

Figure 2

Effect of DGAT inhibition on lipid droplets and total cellular triacylglycerol content. Human myotubes were grown and differentiated on glass bottom microwell dishes. At day 7 of differentiation, myotubes were incubated with 100 µM OA for 4 (A–C) and 24 h (D) in presence or absence of DGAT inhibitors; D1i (1 µM) and D2i (10 µM). The cells were stained for lipid droplets (green) and nuclei (blue) using Bodipy 493/503 and Hoechst 33528, respectively, and images taken with a 60x objective on a confocal microscope. Scale bar 25 µm. (A–C) Representative images are presented for control (A), D1i (B) and D2i (C). (D) LDs were quantified (ImageJ) by relating number of LDs to number of nuclei. Results represent mean ± SEM from one experiment at 4 h and one experiment at 24 h where calculations are based on 6–8 different images for each condition, unpaired t-test, *p < 0.05 vs control at the same time-point. (E) Human myotubes were incubated with 100 µM OA in presence or absence of D1i (1 µM) and D2i (10 µM) for 24 h and total TAG content was measured. The results are presented as mean ± SEM as % of control from n = 3 individual experiments with one 75 cm² cell culture flask for each condition. *p < 0.05 vs. control, paired t-test. D1i, DGAT1 inhibitor; D2i, DGAT2 inhibitor; LDs, lipid droplets; OA, oleic acid; TAG, triacylglycerol.

Figure 3

Effect of DGAT inhibition on incorporation of glycerol, acetate and glucose into TAG and total lipids. Human myotubes were grown and differentiated on 12-well tissue culture plates for 6–7 days. At day 7 of differentiation myotubes were incubated with D-[¹⁴C(U)]glycerol (1 µCi/ml, 10 µM) supplemented with D1i (1 µM) or D2i (10 µM), in presence or absence of 100 µM oleic acid for 4 h (A,D) or with 100 µM [¹⁴C]acetate (2 µCi/ml) in presence or absence DGAT inhibitors for 4 h (B,E). C,F) Cells were incubated with the liver X receptor agonist T0901317 (1 µM) for 96 h. Thereafter, myotubes were incubated for 24 h with D-[¹⁴C(U)]glucose (2 µCi/ml, 5.5 mM) in presence or absence of DGAT inhibitors. Lipids were separated by thin layer chromatography and measured using liquid scintillation. The sum of all radioactivity on the TLC plate is presented as total lipids. Results represent mean ± SEM from n = 4 (A,B), n = 5 (D,E) or n = 3 (C,F) individual experiments with 3 separate culture wells for each condition presented at absolute values (A, B, C) or normalized to control (D–F). *p < 0.05 vs. control, paired t-test. D1i, DGAT1 inhibitor; D2i, DGAT2 inhibitor; TAG, triacylglycerol.

Figure 4

Effect on DGAT inhibition on turnover of accumulated lipids. Human myotubes were grown and differentiated on 96-well tissue culture plates, SPA plates were used for time-course (B,C). At day 6 of differentiation myotubes were incubated with [¹⁴C]oleic acid (0.5 µCi/ml, 100 µM) for 24 h. After 24 h pre-treatment with [¹⁴C]oleic acid myotubes were washed and re-incubated with D1i (1 µM) or D2i (10 µM). (A) Cell-associated radioactivity from [¹⁴C]oleic acid was measured after 4 h treatment with DGAT inhibitors. (B,C) Decline in cell-associated radioactivity was measured over 6 h in presence of an inhibitor of carnitine palmitoyltransferase 1 (etomoxir, 10 µM) and DGAT inhibitors; D1i (1 µM) and D2i (10 µM). Results represent mean ± SEM as nmol/mg protein (A,B) and as all-over effects normalized to control (C) from n = 4 individual experiments with 8 separate culture wells for each condition. *p < 0.05 vs control, paired t-test (A), LMM statistical test (C). D1i, DGAT1 inhibitor; D2i, DGAT2 inhibitor; SPA; scintillation proximity assay; LMM, Linear mixed model.

Figure 5

Effect of DGAT inhibition on oleic acid oxidation. Human myotubes were grown and differentiated on 12- or 24-well tissue culture plates. At day 7 of differentiation, myotubes were incubated with 100 µM [¹⁴C]oleic acid (0.5 µCi/ml, 100 µM) for 4 h in presence or absence of D1i (1 µM) or D2i (10 µM), respectively. Complete oxidation (CO₂ production) and β-oxidation (ASMs) were measured. (A–C) Complete oxidation (CO₂) of oleic acid. (D–F) ASMs, which reflects incomplete oxidation (β-oxidation), of oleic acid. Results are presented as mean ± SEM from n = 5–6 (D1i) and n = 8 (D2i) individual experiments with 8 separate wells for each condition, as absolute values (nmol/mg protein) (A,B and D,E) and normalized to control (C,F). *p < 0.05 vs control, paired t-test. ASMs, acid soluble metabolites; D1i, DGAT1 inhibitor; D2i, DGAT2 inhibitor.

Figure 6

Effect of DGAT inhibitors on glucose metabolism. (A–B) Human myotubes were grown on 96-well tissue culture plates. At day 7, myotubes were incubated with [¹⁴C(U)]glucose (0.5 µCi/ml, 200 µM) and D1i or D2i for 4 h. Oxidation, measured as CO₂ production from [¹⁴C(U)]glucose (A) and cell-associated radioactivity from [¹⁴C(U)]glucose (B) was measured after 4 h treatment with DGAT inhibitors; D1i (1 µM) and D2i (10 µM). (C) Alternatively, cells were grown in 12-well tissue culture plates. At day 7, myotubes were starved for 90 min (DMEM without glucose) before incubation for 3 h with D-[¹⁴C(U)]glucose (1 µCi/ml, 5.5 mM) in presence or absence of D1i (1 µM) or D2i (10 µM) and with or without insulin (100 nM). Results are presented as mean ± SEM from n = 9 (A,B) individual experiments with 8 separate wells for each condition, or n = 4 (C) individual experiments with duplicate wells for each condition. Absolute values for glycogen synthesis: control 11.0 ± 4.5 and with insulin 22.4 ± 10.8 nmol/mg cell protein. *p < 0.05 vs control, paired t-test.

Figure 7

Model illustrating possible functions of DGAT1 and DGAT2 in human skeletal muscle cells. DGAT1 is dominant in human skeletal muscle cells utilizing FAs derived from various sources (exogenously supplied, de novo FA synthesis or FA derived from lipolysis) to generate TAG, both through the monoacylglycerol pathway and the de novo pathway from glycerol-3-P. DGAT2 seems to be specialised only for the synthesis of TAG, involving de novo incorporation of the glycerol moiety into TAG. Moreover, DGAT1 seems to operate at higher substrate concentrations, whereas DGAT2 may esterify substrates at lower concentrations for storage in TAG. Fatty acid oxidation from exogenous FA is regulated by DGAT activities. DAG, diacylglycerol; DGAT, diacylglycerol acyltransferase; FA, fatty acid; Glycerol-3-P, glycerol-3-phosphate; MAG, monoacylglycerol; TAG, triacylglycerol; TAG′ and TAG″, heterogeneous pools of TAG.