Exercise increases TBC1D1 phosphorylation in human skeletal muscle

Subjects.

Ten obese but otherwise healthy subjects (7 women, 3 men) without any ongoing pharmacological treatment except for oral contraceptives were recruited for the study. Subject characteristics and prediet food intake are given in Table 1. All potential participants underwent a comprehensive medical examination as well as routine blood sample tests to exclude any unknown medical conditions, including type 2 diabetes. None of the subjects were performing any strenuous physical exercise on a regular basis. The study was approved by the regional ethics committee in Stockholm, Sweden, and all participants gave their written informed consent after being informed about the purpose and nature of the study.

Study protocol.

All subjects underwent a 2-wk period of dietary intervention, as described below. Before and after the diet intervention, subjects underwent assessment of maximal oxygen uptake (V̇o_{2 max}), body composition (dual-energy X-ray absorption), and insulin sensitivity (hyperinsulinemic clamp). On a separate occasion before and after the dietary intervention, a muscle specimen was obtained from the vastus lateralis muscle at basal and after a 30-min period of cycle ergometer exercise at a workload corresponding to 70% of V̇o_{2 max}.

Exercise protocol and muscle biopsies.

Maximal oxygen consumption (V̇o_{2 max}) was determined on a cycle ergometer, using a stepwise load increase protocol. O₂ uptake, ventilatory volume, and CO₂ production were measured on line (Sensormedics V_max 229), and the leveling-off criterion was used to define V̇o_{2 max} (38).

In the morning on the day of the exercise protocol, a basal muscle specimen was obtained after a 30-min period of rest. A 5- to 7-mm incision was made using local anesthesia and sterile conditions 12–15 cm proximal to the superior border of the patella, and a muscle specimen was obtained from the superficial border of the vastus lateralis muscle using the Bergstrom needle. This was followed by a 30-min period of exercise on the cycle ergometer at a workload corresponding to 70% of determined V̇o_{2 max}, after which another muscle specimen was obtained from the contralateral leg following the same procedure. The muscle tissue was immediately dissected free from fat and connective tissue and placed in liquid nitrogen.

Hyperinsulinemic normoglycemic clamp.

The subjects reported to the laboratory after an overnight fast. After basal blood sampling, a hyperinsulinemic normoglycemic glucose clamp was started. Repeated blood sampling was performed during the steady-state period of the clamp, as described below.

Normoglycemic hyperinsulinemia was created by continuous infusion of insulin (Actrapid; NovoNordisk, Bagsvaerd, Denmark), which was given at a rate of 1.0 mU·kg⁻¹·min⁻¹, aiming to yield a standardized plasma insulin concentration. The blood glucose concentration was kept at normoglycemic levels (as close as possible to 4.5 mM) by controlling blood glucose levels at least every 10 min by adjusting the flow rate of a variable 200 mg/ml glucose infusion (Braun, Melsungen, Germany). After insulin had been infused for ∼60 min, steady-state glucose concentrations were achieved, and the amount of glucose infused during the following 60 min was used to calculate whole body insulin sensitivity (mg glucose·kg body wt⁻¹·min⁻¹), as described earlier (6).

Body composition, including body fat and fat-free mass, was measured using dual-energy X-ray absorptiometry (Prodigy Lunar; GE Lunar, Madison, WI).

Dietary intervention.

The subjects were instructed to consume a low-carbohydrate, high-protein, high-fat diet. Each subject met individually before, after 1 wk, and after the dietary intervention period with a registered dietitian to perform a 24-h dietary recall and to review the central features of the diet. The macronutrient content of the diet was calculated using Dietist XP (Bromma, Sweden) software. The diet consisted of limiting carbohydrate intake without restricting consumption of fat and protein. In brief, subjects were instructed to reduce carbohydrate intake to 21 g/day but were allowed to eat protein and fat as much and as often as they wanted. This included fresh fish, eggs, beef, turkey, chicken breasts without skin, fresh ham slices, raw or steamed vegetables, and butter. Only limited amounts of cheese and cream cheese were allowed. In addition, specific brands of salad dressings and snack foods with limited carbohydrate content were allowed, whereas sauces, gravies, or other ingredients that contain carbohydrates were not recommended. The complete diet instructions can be found in the supplemental data. Subjects were instructed to maintain their regular degree of physical exercise during the intervention period.

Polymerase chain reaction.

Total RNA was isolated from human muscle and was reverse-transcribed to cDNA. TBC1D1 was amplified by PCR with splice exon-flanking primers. Primers utilized were as follows: first set, forward 5′-CCATCAGTGTGGATCTGGA-3′ and reverse 5′-GCCCATCTTCACAAACTGGT-3′; second set, forward 5′-TCAGGAGCAGGCGACTATTT-3′ and reverse 5′-CACAGGAGTCCCACCAGAAT-3′. Amplicons were separated by agarose gel electrophoresis and imaged with ethidium bromide staining under UV light.

Measurements of intracellular signaling.

Muscle biopsies were homogenized in a buffer containing 20 mM Tris, 50 mM NaCl, 5 mM Na₄P₂O₇, 50 mM NaF, 250 mM sucrose, 2 mM DTT, 1% Triton-X 100, 2 μg/ml aprotinin, 5 μg/ml leupeptin, 0.5 μg/ml pepstatin, 10 μg/ml antipain, 1.5 mg/ml benzamidine, and 500 μmol/l phenylmethanesulfonylfluoride, pH 7.4. Muscle lysates (200 μg) were immunoprecipitated with specific antibodies to the α1 and α2 catalytic subunits of AMPK and protein A beads (Sigma). The reaction was performed using the synthetic SAMS peptide as substrate, as described previously (20, 39).

Western blot analyses were used to assess protein and phosphorylation levels of various proteins. Antibodies to AMPKα1 and phosphospecific (Ser⁷⁹)-acetyl-CoA carboxylase (ACC) antibody were from Millipore (Billerica, MA). GLUT4 antibody was from Chemicon (Temecula, CA). Phosphospecific AMPK (Thr¹⁷²), Akt, phosphospecific Akt (Ser⁴⁷³ and Thr³⁰⁸), TCB1D1, and AS160 antibodies were from Cell Signaling Technology (Beverly, CA). ACC expression was assessed using HRP-conjugated streptavidin (Pierce Chemical, Rockford, IL). Antibody to AMPKα2 was generated as described previously (20). Phosphospecific TBC1D1 (Ser²³¹, Ser⁶⁶⁰, Ser⁷⁰⁰, Thr⁵⁹¹) and TCB1D4 Ser⁷¹¹ were generated as described previously (35, 36). Blots were developed using ECL reagents (Amersham Pharmacia Biotech, Piscataway, NJ) and quantified using FluorChem version 2.01 (Alpha Innotech, San Leandro, CA).

Glycogen concentrations.

Muscle samples were hydrolyzed in 2 N HCl at 100°C for 2 h followed by neutralization with 2 N NaOH, and glycogen content was measured by the hexokinase enzymatic method using a glucose HK reagent (Eagle Diagnostics, Desoto, TX).

Statistical analysis.

Data are expressed as means ± SE. Normality of the data was tested with the Kolmogorov-Smirnov test of normal distribution. Where P > 0.20, the data was considered to be normally distributed. All normally distributed data were compared using Student's t-test or one- or two-way ANOVA. The differences between groups were considered significant when P < 0.05.

Clinical and metabolic characteristics of the subjects.

The effects of the diet intervention on subject characteristics and food intake are summarized in Table 1. There was a major shift in macronutrient intake during the diet intervention. The percent energy intake from carbohydrates was reduced from 48 ± 8 to 5 ± 1, a 90% reduction. Fat and protein increased from 32 ± 8 and 19 ± 5 to 59 ± 5 and 36 ± 5, respectively (P < 0.05 for all). The caloric intake during the diet intervention was not statistically significant from the prediet. As has been shown in previous studies, this diet causes a rapid weight loss (2). The average weight loss after diet intervention was 4% (4.2 ± 0.4 kg), and this was associated with an increase in insulin sensitivity as indicated by an ∼30% increase in the M value and a small but significant decrease in fasting blood glucose concentrations. The weight loss was composed of a 4% loss of lean body mass and 3% loss of fat mass. There was also a significant decrease in plasma triglyceride concentrations, whereas cholesterol concentrations were unaffected.

Human skeletal muscle expresses multiple splice isoforms of TBC1D1.

Mouse skeletal muscles express the short and long splice isoform of TBC1D1, with the long form predominant (31). Interestingly, only the long form contains the Ser⁶⁶⁰ and Ser⁷⁰⁰ phosphorylation sites, whereas the Ser²³¹ and Thr⁵⁹⁰ sites are expressed in both splice variants (3). It is not known whether multiple splice variants of TBC1D1 are expressed in human skeletal muscle, and therefore, we determined the relative expression of the long and short TBC1D1 splice variants in human skeletal muscle by amplifying TBC1D1 by PCR with splice exon-flanking primers. The amplicons were separated by agarose gel electrophoresis. Two sets of primers each yielded three products (Amplicon DNA level: 1:0.41:0.28) (Supplemental Fig. S1; Supplemental Material for this article is available online at the AJP-Endocrinology and Metabolism website). Sequencing results confirmed that all three amplicons are splice variants of TBC1D1. The short-form TBC1D1 is missing the entire splice exon (SE) domain, whereas the medium form lacks only the NH₂-terminal part of the SE domain in TBC1D1. Our results suggest that three splice variants of TBC1D1 are expressed in human skeletal muscle. The weight of the short form is predicted to be ∼140 kDa, the medium form 146–148 kDa, and the long form ∼155 kDa. As shown in Fig. 1A, TBC1D1 could be detected by a clear band just above the 150-kDa marker, and at longer exposures two faint bands below the 150-kDa marker could be observed. These data suggest that the long form of TBC1D1 is the predominant splice variant expressed in human skeletal muscle. We quantified the predominant TBC1D1 band (Fig. 1B) and AS160 (Fig. 1C) and did not see an effect of diet intervention on protein expression of TBC1D1 and AS160.

Exercise increases site-specific TBC1D1 phosphorylation in human skeletal muscle.

AMPK and Akt are established upstream kinases of TBC1D1 and AS160 in skeletal muscle (15, 31, 34). Both before and after the dietary intervention, 30 min of moderate-intensity exercise significantly increased activity of AMPKα2 (Fig. 2A), the predominant catalytic isoform expressed in human skeletal muscle, but had no effect on AMPKα1 activity (Fig. 2B). The 2-wk diet intervention had no effect on AMPK activity in the basal state or after exercise stimulation (Fig. 2, A and B). Consistent with these data, AMPK Thr¹⁷² phosphorylation (Fig. 2C) and phosphorylation of the AMPK substrate ACC (Fig. 2D) were increased with exercise, with no effect of the dietary intervention. Exercise did not lead to detectable increases in Akt Ser⁴⁷³ and Akt Thr³⁰⁸ phosphorylation before or after the dietary intervention (Supplemental Fig. S2). Thus, 30 min of moderate-intensity exercise in obese individuals increases AMPKα2 activity but has no effect on Akt or AMPKα1 activity. Skeletal muscle protein expression of AMPK (Fig. 2E), ACC (Fig. 2F), Akt, and the AMPK upstream kinase LKB1 (Fig. 2G) was not altered following the 2 wk on the low-carbohydrate diet.

Using phosphoproteomics and immunoblotting, we determined previously that Ser²³¹, Ser⁶⁶⁰, and Ser⁷⁰⁰ on TBC1D1 are phosphorylated in mouse skeletal muscle after contraction and during stimulation with the AMPK activator 5-aminoimidazole-4-carboxamide-1-β-d-ribofuranoside (AICAR) (31, 36). We also identified Thr⁵⁹⁰ as an Akt site that is phosphorylated with insulin stimulation (31, 36). Using site-specific antibodies that recognize TBC1D1 phosphorylation on these residues, we determined regulation of TBC1D1 phosphorylation in human skeletal muscle. Exercise significantly increased Ser²³¹and Ser⁶⁶⁰ phosphorylation, whereas there was only a slight tendency for exercise to increase TBC1D1 Ser⁷⁰⁰ phosphorylation (Fig. 3, A–C). The insulin-responsive TBC1D1 Thr⁵⁹¹ site (Fig. 3D) was not affected by the acute bout of exercise, consistent with the absence of Akt phosphorylation. Diet intervention reduced the basal level of Ser⁶⁶⁰ phosphorylation but did not modify the response to exercise. There was no effect of the low-carbohydrate diet intervention on any of the other TBC1D1 phosphorylation sites under either the basal or exercise condition (Fig. 3, A, C, and D).

We recently identified AS160 Ser⁷¹¹ as an AMPK consensus sequence in mouse skeletal muscle (35). In the current study, 30 min of exercise robustly increased AS160 Ser⁷¹¹ phosphorylation in human skeletal muscle (Fig. 3E). Consistent with the absence of Akt phosphorylation, exercise did not increase AS160 phospho-Akt substrate (PAS) phosphorylation (Fig. 3F). There was no effect of the low-carbohydrate diet intervention on AS160 Ser⁷¹¹ or PAS phosphorylation (Fig. 3, E and F).

Muscle glycogen and GLUT4 concentrations.

Muscle glycogen concentration is a major regulator of skeletal muscle glucose transport (10, 13). The 2-wk diet intervention, which was composed of low carbohydrate consumption, significantly reduced muscle glycogen concentrations (Fig. 4A). The single bout of acute exercise significantly decreased muscle glycogen concentrations both before and after the diet intervention (Fig. 4A). GLUT4 expression was not altered by the dietary intervention (Fig. 4B).