Stearoyl-CoA desaturase (SCD), a rate-limiting enzyme in the biosynthesis of monounsaturated fatty acids, has recently been shown to be a critical control point in regulation of liver and skeletal muscle metabolism. Herein, we demonstrate that endurance training significantly increases both SCD1 mRNA and protein levels in the soleus muscle, whereas it does not affect SCD1 expression in the EDL muscle and liver. Desaturation index (18:1Δ9/18:0 ratio), an indirect indicator of SCD1 activity, was also significantly higher (3.6-fold) in soleus of trained rats compared with untrained animals. Consistent with greater SCD1 expression/activity, the contents of free fatty acids, diacylglycerol, and triglyceride were elevated in soleus of trained rats. However, training did not affect lipid concentration in EDL and liver. Additionally, endurance training activated the AMP-activated protein kinase pathway as well as increased peroxisome proliferator-activated receptor (PPAR)-δ and PPARα gene expression and activity in soleus and liver. Increased lipid accumulation in soleus was coupled with elevated protein levels of fatty acid synthase, mRNA levels of diacylglycerol acyltransferase and glycerol-3-phosphate transferase, as well as increased levels of proteins involved in fatty acid transport (fatty acid translocase/CD36, fatty acid transport protein 1). Interestingly, sterol regulatory element-binding protein (SREBP)-1c expression and SREBP-1 protein levels were not affected by exercise training. Together, the obtained data suggest that SCD1 upregulation plays an important role in adaptation of oxidative muscle to endurance training.
- exercise training
- fatty acid transport proteins
- fatty acid synthase
- sterol regulatory element-binding protein 1
regulation of triglyceride (TG) deposition in the skeletal muscle has drawn a lot of attention recently because increased accumulation of intramuscular TG (IMTG) is thought to be related to insulin resistance (2, 22). Interestingly, despite enhanced skeletal muscle insulin sensitivity and oxidative capacity, highly trained athletes also have increased IMTG (22, 47). Furthermore, moderate exercise training [1–1.5 h of cycle exercise at 45–65% of maximal oxygen consumption (V̇o2max)] in previously sedentary subjects increases IMTG level (43, 46). The mechanism controlling increase in IMTG in exercise training has not been fully determined. One possibility is the upregulation of the sterol regulatory element-binding protein 1c (SREBP-1c), a transcription factor that regulates expression of genes involved in the biosynthesis of cholesterol, triglycerides (TG), fatty acids (FA), and phospholipids (PL) (21, 36). Another option has been proposed recently by Schenk and Horowitz (43), who showed that a single session of exercise increases the abundance of proteins involved in TG synthesis, i.e., mitochondrial glycerol-3-phosphate transferase (GPAT), diacylglycerol acyltransferase 1 (DGAT1), and stearoyl-CoA desaturase (SCD1) in human skeletal muscle, thereby possibly increasing IMTG synthesis. It is interesting to note that overexpression of either GPAT (20), DGAT (33), or SCD1 (30) significantly increases TG synthesis in numerous cell types. However, the role of SCD1 in lipid biosynthesis and the accumulation of FA metabolites during endurance training is not known.
Recent studies have shown that SCD1, the rate-limiting enzyme catalyzing the biosynthesis of monounsaturated fatty acids, is a principal regulator of lipogenesis (12). SCD1 deficiency leads to a decrease in the content of hepatic TG and cholesteryl esters (ChE) (10, 34, 48) and downregulates de novo FA synthesis (8, 33) and lipoprotein secretion (25) in liver. In contrast, overexpression of SCD1 in CHO cells results in a significant increase in the rate of TG synthesis (30). A similar relationship between SCD1 expression and TG accumulation was also found in skeletal muscle (8, 11). Lack of SCD1 gene increases the rate of FA β-oxidation through activation of the AMP-activated protein kinase (AMPK) pathway and by upregulating genes of fatty acid oxidation in soleus and red gastrocnemius muscles (8, 9). Consistent with increased β-oxidation, the contents of free fatty acids (FFA) and long-chain acyl-CoAs are significantly lower in the soleus and red gastrocnemius of SCD1−/− mice (9). SCD1 deficiency also reduces ceramide biosynthesis in oxidative myofibers in both wild-type and leptin-deficient ob/ob mice (8) and protects against lipid-induced insulin resistance (11, 40). On the other hand, Doran et al. (13) showed that IMTG accumulation caused by a reduced-protein diet is coupled with increased SCD expression and activity as well as elevated monounsaturated and total FA levels in muscle (13). Surprisingly, conjugated linoleic acid, which was shown to decrease SCD1 gene expression in 3T3-L1 adipocytes (4) and in liver (28), has no significant effect on SCD activity in muscle (7, 23). Furthermore, Ikeda and coworkers (21) showed that 2 wk swimming increases SREBP1 mRNA level in skeletal muscles (gastrocnemius, quadriceps), which led to IMTG accumulation coupled with an increase in expression of lipogenic genes, including SCD1. Similar effects were found in muscle of trained humans (2).
Based on the information presented above, we hypothesized that SCD1 may play an important role in the partitioning of excess FA toward IMTG synthesis observed in skeletal muscles during long-term exercise. We show here that 6 wk of endurance training increases FFA, diacyloglycerol (DAG), and TG accumulation in the soleus muscle, whereas the lipid content in extensor digitorum longus (EDL) and liver is not affected. The changes in lipid concentration in soleus are parallel with increased mRNA expression and protein level of SCD1. Also the desaturation index (18:1Δ9/18:0), an intermediate indicator of SCD1 activity, is more than 3.5-fold increased in soleus in trained rats compared with untrained animals. Endurance training does not affect SCD1 expression or the desaturation index in EDL and liver. The obtained results show that SCD1 upregulation is coupled with IMTG accumulation occurring after endurance exercise.
MATERIALS AND METHODS
Twenty-four male Wistar rats (body wt 220 ± 10 g) were randomly assigned to either a treadmill running program (n = 12) or served as sedentary controls (n = 12). The animals were fed standard laboratory CHOW diet (5008 test diet; PMI Nutrition International, Richmond, IN) and had free access to food and water. Rats to be exercise trained were subject to running exercise 5 days/wk for 6 wk, beginning at a treadmill speed of 16 m/min (0° inclination). The running speed was increased by 4 m/min per week over the first 4 wk, and then maintained at 28 m/min for the remaining 2 wk of the endurance training. The running time was 40 min/day over the first 4 wk, and next was daily increased by 5 min to reach 60 min (26). The relative workload during this type of endurance training was 50–60% V̇o2max (44). This training protocol results in a significant increase in the anaerobic threshold, expressed as the running speed at which lactate starts to accumulate in the blood, from 23 m/min in the control group to 29.5 m/min in the trained group. Anaerobic threshold shifted toward higher workload in endurance-trained animals implies a bigger reliance of skeletal muscle on fat metabolism (26). Also, the activity of 2-oxoglutarate-dehydrogenase, a regulatory enzyme of the tricarboxylic acid (TCA) cycle, was significantly increased (data not shown), additionally confirming an enlarged oxidative capacity induced by the employed training. The endurance-trained rats were killed 48 h after the last exercise session by decapitation. Liver, soleus, and EDL muscles were excised, cleaned of any visible adipose tissue, nerves, and fascias, and frozen in liquid nitrogen. The experiments were approved by the Ethical Committee for Animal Experiments at the Mossakowski Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland.
SCD1, fatty acid synthase (FAS), fatty acid translocase/CD36 (CD36), fatty acid transport protein 1 (FATP1), SREBP-1, and β-actin antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). AMPK and phospho-AMPK (pAMPK) antibodies were from Cell Signaling (Hartsfordshire, UK), whereas phospho-acetyl-CoA carboxylase (ACC) antibody was from Upstate (New York, NY). Horseradish peroxidase-conjugated streptavidin was from Pierce (Rockford, IL). Other chemicals were purchased from Sigma (St. Louis, MO).
Isolation and analysis of RNA.
Total RNA was isolated from tissues of trained and control rats using TRIzol reagent (Invitrogen, Carlsbad, CA). DNase-treated RNA was reverse transcribed with Superscipt III (Invitrogen) and real-time quantitative PCR was performed on an ABI Prism 7500 Fast Instrument. SYBR green was used for detection and quantification of given genes expressed as mRNA level normalized to a gene encoding actin using the ΔΔCt method. Relative abundance of SCD1, SREBP-1c, peroxisome proliferator-activated receptor-α (PPAR)-α, PPARδ, acyl-CoA oxidase (ACO), carnitine palmitoyltransferase 1 (CPT1), PPAR-γ coactivator-1α (PGC1α), uncoupling protein 2 (UCP2), DGAT1, DGAT2, and GPAT was measured. Primer sequences are available on request.
Measurement of lipids.
Lipids were extracted by the method of Bligh and Dyer (3) and measured as described (9). Briefly, the lipids were separated by thin-layer chromatography (TLC) on silica gel-60 plates (Merck, Darmstad, GE) in heptane/isopropyl ether/glacial acetic acid (60/40/4, vol/vol/vol) with authentic standards. The bands corresponding to TG, FFA, DAG, and PL standards were scraped off the plate and transferred to screw cap glass tubes containing methylpentadecanoic acid as an internal standard. FA were then transmethylated in the presence of 14% boron trifluoride in methanol. The resulting methyl esters were extracted with hexane and analyzed by gas-liquid chromatography (GC). Total lipid contents were calculated from the individual FA content in each fraction.
The content of oleic (18:1Δ9) and stearic (18:0) acids in total lipid extracts was analyzed by GC, as described above, and used for calculation of the 18:1Δ9/18:0 ratio.
Western blot analysis.
The samples were homogenized and centrifuged at 10,500 g for 20 min in ice-cold 50 mM HEPES buffer (pH 7.4) containing 150 mM NaCl, 10 mM sodium pyrophosphate, 2 mM Na3VO4, 10 mM NaF, 2 mM EDTA, 2 mM PMSF, 5 μg/ml leupeptin, 1% Nonidet P-40, and 10% glycerol. Protein levels of SCD1, FAS, CD36, FATP1, SREBP-1, and α1- and α2-AMPK, and the extent of phosphorylation of AMPK at Thr172 and of ACC at Ser79 were determined in 50 μg of clarified homogenate protein using specific antibodies. The separated proteins (9% SDS-PAGE gels) were transferred to PVDF membranes (Millipore, Billerica, MA), which were blotted using appropriate antibodies. To measure ACC protein level membranes were incubated for 1 h in streptavidin-horseradish peroxidase (Pierce, Rockford, IL) (24). The proteins were visualized using ECL (Pierce) as described by the manufacturer and quantified by densitometry. Protein level of SCD1, FAS, CD36, FATP1, and SREBP-1 was expressed relative to β-actin abundance, whereas phosphorylation of AMPK and ACC was expressed relative to the abundance of the respective proteins.
Protein concentration was determined with Bio-Rad protein assay (Bio-Rad, Hercules, CA) using BSA as a standard.
Results were analyzed using the Student's t-test. A difference of P < 0.05 was considered significant. Values are presented as means ± SD (n = 12 rats per group).
Increased phosphorylation of AMPK and ACC in liver and soleus after endurance training.
AMPK is activated by phosphorylation at a threonine residue (17). We therefore measured AMPK phosphorylation and AMPK α-subunit protein level in tissue homogenates of trained and untrained rats. AMPK α-subunit protein levels were not affected by exercise training in liver, soleus, and EDL (Fig. 1A). However, phosphorylation of AMPK α-subunit was significantly increased in soleus (by 78%) and in liver (by 71%), whereas it was not changed in EDL (Fig. 1A).
Phosphorylation of ACC at Ser79 by AMPK leads to inhibition of ACC activity (6). Consistent with greater AMPK phosphorylation observed after training, ACC phosphorylation was also significantly increased in soleus and liver (by 90% and 55%, respectively), whereas it remained unchanged in EDL (Fig. 1B). The ACC protein level was not affected by training in any of the investigated tissues (Fig. 1B).
PPARδ and PPARα pathways are upregulated in liver and soleus after long-term exercise.
PPARδ and PPARα are important metabolic regulators in skeletal muscle. Their basic function relates to transcriptional programs that enhance fatty acid catabolism and energy uncoupling (16). The mRNA levels of PPARδ and of PPARα were significantly increased in liver and soleus of trained animals compared with untrained rats, whereas expression of these genes was not affected by training in EDL (Fig. 2, A and B). To assess the impact of exercise training on PPAR pathways activation, we measured expression of genes that are regulated by PPARδ and PPARα: ACO, CPT1, PGC1α, and UCP2. The relative expression of these genes was increased significantly in liver (except for UCP2) and soleus of trained animals compared with untrained rats, whereas expression of these genes was not changed by training in EDL (Fig. 2, C–F). Taken together, these data suggest that activation of the AMPK- and PPAR(s)-pathways may be accountable for increased oxidative capacity induced by endurance training.
Endurance training increases lipid content in soleus muscle.
Exercise training did not affect FFA, DAG, TG, and PL concentrations in liver and EDL (Fig. 3). However, in soleus, training significantly increased the content of FFA (109.78 ± 3.3 in control vs. 447.53 ± 71.1 nmol/g in trained rats), DAG (421.59 ± 104.8 in control vs. 708.57 ± 55.4 nmol/g in trained rats), and TG (8.87 ± 2.1 in control vs. 34.65 ± 3.98 μmol/g in trained rats), whereas the level of PL after exercise training was only slightly elevated (14.87 ± 2.1 in control vs. 17.03 ± 2.42 μmol/g in trained rats) (Fig. 3). Increased contents of FFA, DAG, and TG in soleus of trained animals were accompanied by elevated levels of unsaturated FA in each of these lipid fractions (data not shown).
SCD1 expression is increased in soleus after exercise training.
To assess the link between SCD1 expression and lipid synthesis in the skeletal muscle and liver we measured SCD1 mRNA and protein level by real-time PCR and Western blot, respectively. Consistent with increased lipid content in soleus, both SCD1 mRNA and protein levels were increased in this muscle (by 3.2-fold and 1.8-fold, respectively; Fig. 4, A and B). Interestingly, neither SCD1 mRNA nor SCD1 protein level was changed after training in EDL and liver (Fig. 4, A and B).
Desaturation index is an indirect indicator of desaturase activity (1). We used the product-to-precursor ratio (18:1Δ9/18:0) calculated from the content of respective fatty acids in lipids extracted from soleus, EDL, and liver. The desaturation index was not significantly changed due to exercise training in EDL and liver (Fig. 4C). However, in the soleus muscle the 18:1Δ9/18:0 ratio was significantly increased (3.6-fold) in exercised animals compared with the untrained group (Fig. 4C), indicating higher SCD1 activity in this tissue.
Endurance training increases FAS protein level and lipogenic genes expression in soleus.
Another important lipogenic enzyme that catalyzes de novo fatty acid synthesis is FAS. Similarly as in the case of SCD1, FAS protein level was significantly increased after training in the soleus muscle but remained unchanged in EDL and liver compared with untrained rats (Fig. 5A). Gene expression of GPAT and DGAT, two key enzymes of DAG and TG synthesis (43), was also increased in soleus of exercised rats whereas endurance training did not affect DGAT and GPAT mRNA levels in EDL (Fig. 5B).
Exercise training increases expression of proteins involved in FA transport in soleus.
CD36 and FATP are the major proteins responsible for membrane FA transport (5, 15). Because endurance training increased lipid accumulation in soleus we measured both CD36 and FATP1 protein levels by Western blotting. The content of both proteins was significantly increased in soleus of exercised animals compared with the untrained group (Fig. 6, A and B), indicating that increased FA transport may account for increased lipid synthesis and FA utilization in this muscle. Another proposed mechanism leading to TG accumulation in the skeletal muscle involves upregulation of SREBP-1c (36). However, the 6-wk endurance training did not cause any changes in SREBP-1c gene expression or SREBP-1 mature (60 kDa) and precursor (130 kDa) protein levels (Fig. 6, C and D) in soleus.
SCD1 has been recently shown to be essential for lipid synthesis in animal and cell culture models (12, 39) and was suggested to be one of the factors influencing TG content in human skeletal muscle after exercise (2, 43). In the present study, we established that a 6-wk-long moderate-intensity endurance training increases both SCD1 mRNA and protein levels in the soleus muscle, whereas it does not affect SCD1 expression in the EDL muscle and liver. Desaturation index (18:1Δ9/18:0 ratio) was also significantly higher in soleus after endurance training, indicating increased SCD activity. We did not find increased desaturation index/SCD activity in EDL and liver of trained rats. SCD1 expression has been shown to decrease with exercise in liver (49). However, in that study, despite reduced hepatic SCD1 expression, endurance training did not affect desaturation index or TG accumulation in liver (49). This is in agreement with our present findings. Also, increased SCD1 expression in trained muscles has been previously reported in humans (2) and mice (21); however these studies did not consider muscle fiber types.
In our study, elevated SCD1 expression/activity in trained soleus was coupled with significantly increased contents of FFA, DAG, and TG. In each of these fractions we found increased content of unsaturated FA. Furthermore, endurance training did not affect lipid concentration or unsaturated FA levels in EDL and liver. These data suggest that exercise-induced changes in fat accumulation may be associated with SCD activity. Muscle TG synthesis is likely dependent on the coordinated expression and upregulation of a variety of enzymes (43). A study by Liu et al. (29) demonstrated that short-term exercise (1 wk) increased DGAT activity and TG concentration in soleus. In our study 6-wk-long endurance training also led to an increase in both GPAT and DGAT gene expression in soleus. Studies investigating the effects of unsaturated fatty acids (27, 30, 38) and SCD1 (30, 38) on TG storage clearly demonstrate that the synthesis of TG is increased by the presence of unsaturated FA, or by converting saturated FA to unsaturated ones. Thus increased SCD activity due to endurance training suggests a unique model of muscle TG synthesis regulation that is activated by exercise.
The importance of SCD for TG synthesis may be associated with its subcellular localization. Enzymes involved in TG biosynthesis such as DGAT and microsomal GPAT as well as SCD are located in the endoplasmic reticulum (ER) membrane (10, 31). A possible physiological explanation for the requirement of SCD activity in the synthesis of TG is the production of more easily accessible monounsaturated FA within the vicinity of DGAT and GPAT. In fact, SCD1 was shown to colocalize with DGAT2 in the ER (31), and it is suggested that these two enzymes exist in a complex promoting substrate channeling and fuel partitioning into various metabolic pathways. There is a possibility that monounsaturated FA are incorporated into TG, which are then immediately hydrolyzed, and the FA are channeled to mitochondrial β-oxidation (39). If this hypothesis is true, increased SCD activity may also be involved in processes leading to increased FA utilization in muscle occurring in response to endurance training. Additionally, we showed that training-induced upregulation of SCD is accompanied by increased expression of FAS, the rate-limiting enzyme in biosynthesis of long-chain FA (32). Since the predominant product of FAS (i.e., palmitic acid) is a preferred substrate of SCD, increased FAS expression may be indirectly accountable for increased SCD activity. Parallel upregulation of SCD1 and FAS in soleus after training suggests that these proteins potentially cooperate in maintaining increased monounsaturated FA content and may play a role in increased lipid accumulation induced by exercise. Altogether, this suggests that endurance training not only upregulates the TG synthesis through activation of DGAT and GPAT but also amplifies the capacity for synthesizing TG by upregulating SCD1 and thus by increasing the availability of unsaturated FA.
It is well-known that exercise increases utilization of lipids in skeletal muscles and that replenishment of IMTG occurs after the exercise (22). Our study showed that lipid accumulation triggered by endurance training is tissue and muscle fiber type specific. We did not observe any effect of training on the lipid content in liver and EDL, whereas the contents of TG, DAG, and FFA were significantly increased in soleus of trained rats when compared with sedentary controls. This is probably due to different metabolic needs of these tissues. The soleus is an oxidative muscle that relies on TG as the energy source (22). Thus lipid accumulation in soleus may be an adaptation to increased energetic needs caused by endurance training. This conclusion is confirmed by increased activity of the AMPK, PPARδ, and PPARα pathways observed in soleus after exercise training. Furthermore, these data suggest that the AMPK and PPAR pathways may be involved in molecular mechanisms responsible for increased muscle oxidative capacity induced by endurance training (16, 47). Also an increased protein level of CD36 and FATP1 found after training could contribute to greater availability of FA for β-oxidation in mitochondria. As shown by Nickerson and coworkers (37) overexpession of these proteins in skeletal muscle leads to enhanced mitochondrial FA oxidation but not esterification into TG. These findings are in agreement with results obtained from studies of oxidative skeletal muscle after acute exercise or endurance training (14, 18, 19). In contrast, EDL is a glycolytic muscle and uses glycogen and phosphocreatine instead of TG as a preferential energetic fuel (22). Therefore, in our study, similarly to data obtained by Durante et al. (14) and Nadeau et al. (36), there were no significant differences in AMPK and ACC phosphorylations in EDL muscle. We also observed activation of the AMPK, PPARδ, and PPARα pathways after endurance exercise in liver. It was suggested that elevated AMPK activity in liver may play an important role in adaptation to prolonged exercise (41, 45). However, the physiological relevance of AMPK activation and increased FA oxidation after long-term exercise in liver remains unknown. AMPK has been shown to block TG synthesis in isolated mouse soleus (35). However, our study showed increased AMPK pathway activity in soleus of exercised animals coupled with greater muscle TG accumulation and elevated lipogenic genes expression. These data demonstrate that the mechanism accountable for increased TG synthesis 48 h after exercise is independent of AMPK and ACC activities and relies mainly on upregulation of lipogenesis.
Very little is known about mechanisms involved in regulation of gene expression after exercise training. Nadeau et al. (36) showed that SREBP-1c, a key transcription factor of liver lipogenesis, is upregulated in the gastrocnemius and soleus muscles after 20 wk of treadmill exercise. It was proposed that augmentation of SREBP-1c expression might be responsible for skeletal muscle TG synthesis after chronic exercise training. However, we did not find any differences in SREBP-1c mRNA or protein levels in soleus after 6 wk of endurance training. It was shown that a single bout of exercise can increase SREBP1 protein maturation (43). Therefore, to avoid or at least minimize the effect of the last exercise session, we performed all experiments 48 h after exercise. Our study showed that 6 wk of exercise is insufficient to induce a stable SREBP-1c overexpression or that the molecular mechanisms controlling lipogenesis depend on the duration and/or type of exercise. It was recently reported that gene expression of PPARγ is significantly increased in muscles of endurance-trained athletes (2). It is thus possible that activation of PPARγ may be involved in upregulation of lipogenesis induced by long-term exercise.
In conclusion, after endurance training we observed upregulation of SCD1 coupled to lipid accumulation in the soleus muscle, whereas in the EDL muscle and liver, this effect was not present. Increased SCD1 expression in soleus was coupled with a greater mRNA level of enzymes catalyzing FA and TG biosynthesis. Our study demonstrates that endurance training not only upregulates the TG synthesis through activation of DGAT and GPAT but also amplifies the capacity for synthesizing TG by upregulating SCD. Furthermore, higher SCD activity in soleus was accompanied by increased AMPK and ACC phosphorylation, activation of the PPARδ and PPARα pathways, and by elevated levels of proteins involved in FA transport (CD36, FATP1). These changes were associated with training-induced increase in oxidative capacity and a bigger reliance of skeletal muscle on fat metabolism. A number of recent studies suggest that inhibition of SCD could be a new therapeutic strategy in the treatment of lipid-induced insulin resistance in muscle (11). However, our study demonstrates that SCD1 overexpression is required for muscle TG synthesis induced by endurance training and suggests that SCD plays an important role in adaptation of the oxidative muscle to long-term exercise. If the above is true, diminishing SCD activity in insulin-resistant muscle may be associated with a high risk of muscle dysfunction and should be carefully evaluated.
This work was supported by Polish Ministry of Science and Higher Education Grant No. N N301 0402 36 (to P. Dobrzyn), EMBO Installation Grant No. 1643 (to A. Dobrzyn), and statutory funds from the Nencki Institute of Experimental Biology.
No conflicts of interest, financial or otherwise, are declared by the author(s).
- Copyright © 2010 the American Physiological Society