Article pubs.acs.org/jpr
Insulin Increases Phosphorylation of Mitochondrial Proteins in Human Skeletal Muscle in Vivo Xiaolu Zhao,†,‡ Steffen Bak,†,§ Andreas J. T. Pedersen,∥ Ole Nørregaard Jensen,† and Kurt Højlund*,§,∥,⊥ †
Department of Biochemistry and Molecular Biology, University of Southern Denmark, DK-5230 Odense M, Denmark College of Life Science, Wuhan University, Wuhan, P. R. China 430072 § Section of Molecular Diabetes & Metabolism, Institute of Clinical Research and Institute of Molecular Medicine, University of Southern Denmark, DK-5000 Odense C, Denmark ∥ Department of Endocrinology, Odense University Hospital, DK-5000 Odense M, Denmark ⊥ Section of Molecular Physiology, The August Krogh Centre, Department of Nutrition, Exercise and Sports, University of Copenhagen, 2100 Copenhagen, Denmark ‡
S Supporting Information *
ABSTRACT: There is increasing evidence that multiple proteins involved in key regulatory processes in mitochondria are phosphorylated in mammalian tissues. Insulin regulates glucose metabolism by phosphorylation-dependent signaling and has been shown to stimulate ATP synthesis in human skeletal muscle. Here, we investigated the effect of insulin on the phosphorylation of mitochondrial proteins in human skeletal muscle in vivo. Using a combination of TiO 2 phosphopeptide-enrichment, HILIC fractionation, and LC− MS/MS, we compared the phosphoproteomes of isolated mitochondria from skeletal muscle samples obtained from healthy individuals before and after 4 h of insulin infusion. In total, we identified 207 phosphorylation sites in 95 mitochondrial proteins. Of these phosphorylation sites, 45% were identified in both basal and insulin-stimulated samples. Insulin caused a 2-fold increase in the number of different mitochondrial phosphopeptides (87 ± 7 vs 40 ± 7, p = 0.015) and phosphoproteins (46 ± 2 vs 26 ± 3, p = 0.005) identified in each mitochondrial preparation. Almost half of the mitochondrial phosphorylation sites (n = 94) were exclusively identified in the insulin-stimulated state and included the majority of novel sites. Phosphorylation sites detected more often or exclusively in insulin-stimulated samples include multiple sites in mitochondrial proteins involved in oxidative phosphorylation, tricarboxylic acid cycle, and fatty acid metabolism, as well as several components of the newly defined mitochondrial inner membrane organizing system (MINOS). In conclusion, the present study demonstrates that insulin increases the phosphorylation of several mitochondrial proteins in human skeletal muscle in vivo and provides a first step in the understanding of how insulin potentially regulates mitochondrial processes by phosphorylation-dependent mechanisms. KEYWORDS: mitochondria, insulin stimulation in vivo, oxidative phosphorylation, human skeletal muscle, phosphoproteomics
■
INTRODUCTION Mitochondria are the primary energy-generating system in mammalian cells and play a central role in numerous metabolic processes such as oxidative phosphorylation (OxPhos), the tricarboxylic acid (TCA) cycle, β-oxidation, and amino acid metabolism, as well as apoptosis and calcium homeostasis.1 Abnormalities in mitochondrial function have been implicated in a wide range of common diseases including diabetes, various cancers, neurodegenerative diseases, cardiovascular disease, and aging.2−4 There is emerging evidence that multiple proteins involved in a variety of key regulatory processes in mitochondria are phosphorylated in different mammalian tissues and potentially regulated by specific kinases and phosphatases, which have been localized to mitochondria.5−8 In human vastus lateralis muscle, we recently reported 155 phosphorylation sites in 77 mitochondrial proteins involved in © 2014 American Chemical Society
most known processes in mitochondria in the resting, basal state.9 The phosphorylation of mitochondrial proteins in skeletal muscle may change significantly in response to different stimuli such as changes in circulating hormones and peptides and exercise/contraction. Skeletal muscle is rich in mitochondria and accounts for ∼25% of whole body oxygen consumption in humans in the resting state.10 Moreover, skeletal muscle is the major site of glucose uptake (∼80%) in response to insulin and correspondingly is an important site of insulin resistance in obesity, polycystic ovary syndrome, and type 2 diabetes.11,12 Impaired activation of enzymes in the insulin signaling cascade has been demonstrated in vastus lateralis muscle in these insulin-resistant Received: November 25, 2013 Published: March 18, 2014 2359
dx.doi.org/10.1021/pr401163t | J. Proteome Res. 2014, 13, 2359−2369
Journal of Proteome Research
Article
conditions.11−15 Within the past decade, several studies have provided evidence for a link between insulin resistance and abnormalities in mitochondrial oxidative metabolism in human skeletal muscle (vastus lateralis) even before the development of type 2 diabetes.2,16−18 A recent report of mitochondrial dysfunction in individuals with insulin receptor mutations implies a role for insulin signaling in mitochondrial function19 and suggests that mitochondrial dysfunction is an immediate consequence of insulin resistance. Insulin is known to regulate glucose and lipid metabolism, protein synthesis, and gene expression through phosphorylation-dependent signaling.20 Some studies also indicate that insulin stimulates ATP synthesis in human skeletal muscle and that this response may be attenuated in insulin-resistant individuals.2,21,22 At the molecular level, we have shown that ATP synthase β-subunit is phosporylated at several sites in human skeletal muscle, that one of these sites is regulated by insulin, and that phosphorylation of ATP synthase β-subunit is altered in obesity and type 2 diabetes.23 Moreover, studies of human and rat cell lines and mice myocardium have shown that Akt in its phosphorylated form is translocated into mitochondria in response to insulin and that this is associated with an increased ATP synthase activity.24,25 On the basis of these observations, we hypothesized that insulin regulates mitochondrial function by phosphorylation-dependent mechanisms and hence stimulation of human skeletal muscle with insulin in vivo will cause changes in the phosphorylation of mitochondrial proteins. A first step in understanding how insulin potentially regulates mitochondrial processes is to compare the phosphoproteomes of isolated muscle mitochondria in the basal and insulinstimulated states, respectively. In the present study, we used a strategy of discovery-mode phosphoproteomics combining phosphopeptide enrichment with hydrophilic interaction liquid chromatography (HILIC) fractionation and high-performance tandem mass spectrometry (MS/MS) analysis to study the changes in the mitochondrial phosphoproteome upon in vivo insulin stimulation of human skeletal muscle.
■
lateralis muscle before and after the 4 h of insulin infusion using a modified Bergströ m needle with suction under local anesthesia (10 mL lidocaine 2%). Muscle samples were immediately blotted free of blood, fat, and connective tissue and were used for isolation of mitochondria as described below. The samples were transported in ice cold isolation buffer containing both phosphatase and kinase inhibitors (see below). Materials
Pure water was obtained from a Milli-Q system (Millipore, Bedford, MA). The Microcon YM-10 filter unit was from Millipore (Millipore, Bedford, MA). Endoproteinase Lys-C was from Wako Chemicals USA, Inc.. Modified trypsin was from Promega (Madison, WI). Titanium dioxide (TiO2) resin (5 μm) was obtained from a disassembled titanium dioxide cartridge purchased from GL Sciences Inc. (Tokyo, Japan). TSKGel Amide-80 3 μm HILIC particles were from Tosoh Biosciences. Poros Oligo R3 reversed-phase material was from PerSeptive Biosystems (Framingham, MA). GELoader tips were from Eppendorf (Hamburg, Germany). 3M Empore C8 disks were from Bioanalytical Technologies (St. Paul, MN). All other reagents and solvents were of the highest commercial quality and were used without further purification. Preparation of Mitochondria
Isolation of human skeletal muscle mitochondria was performed as described previously with minor modifications.9 This method yields a high quality of mitochondrial preparations for measurements of respiration in functionally intact mitochondria.26 The muscle samples were cut into very small pieces with scissors and rinsed carefully in isolation buffer (100 mM sucrose, 100 mM KCl, 50 mM Tris-HCl, 1 mM KH2PO4, 0.1 mM EGTA, 0.2% BSA, protein phosphatase inhibitor tablets (PhosphoSTOP Roche, Heidelberg, Germany), protease inhibitor tablet (Roche), 2 mM sodium pervanadate, 1 mM EDTA, pH 7.4). Muscle pieces were then incubated in isolation buffer deprived of protease inhibitor and supplemented with 0.2 mg/mL Nagarse (Sigma) for 2 min. After homogenization using a motor-driven Plexiglas homogenizer, the homogenate was centrifuged at 700 × g for 10 min, and the supernatant was collected and centrifuged at 10,000 × g for 10 min. The pellet was washed, resuspended with isolation medium, and centrifuged at 7,000 × g for 4 min. The final mitochondrial pellet was solubilized and digested as described below. To avoid the dephosphorylation of sites during the isolation procedure, PhosphoSTOP and sodium pervanadate were included in all buffers to completely block the action of any phosphatases in the mitochondrial preparations until lysis with SDS.
METHODS
Subjects and Muscle Biopsies
Skeletal muscle samples were obtained from four healthy, nondiabetic, non-obese volunteers (age 33−57 years; body mass index 24.5−26.0 kg/m2). The volunteers had no family history of type 2 diabetes and were not taking any medication. Informed consent was obtained from all individuals before participation. The study was approved by the Local Ethics Committee and was performed in accordance with the Helsinki Declaration. After an overnight fast, the subjects underwent a 4h euglycemic−hyperinsulinemic clamp using an insulin infusion rate of 80 mU/min/m2.14 We used a 4-h insulin infusion period to be sure that the results could be compared with future studies of patients with type 2 diabetes, in which 4 h of insulin infusion is necessary to obtain steady-state levels of insulinstimulated glucose uptake during the last hour of the clamp.14 The insulin-stimulated glucose infusion rate was calculated as the average amount of glucose (mg/min/m2) needed to maintain euglycemia (∼5.5 mmol/L) during the last 30 min of the clamp. High physiological hyperinsulinemia at ∼900 pmol/ L was obtained in all subjects during the insulin-stimulated periods. The glucose-infusion rates needed to maintain eyglycemia at the end of the clamp ranged from 217 to 556 mg/min/m2. Muscle biopsies were obtained from the vastus
Extraction and Digestion of Mitochondrial Proteins
Mitochondrial proteins were extracted and digested using the filter aided proteome preparation (FASP) control as previously described with minor modifications.27 The mitochondrial pellet was lysed in SDT-lysis buffer (4% w/v SDS, 100 mM Tris/HCl pH 7.6, 0.1 M DTT) at 80 °C for 15 min. The protein concentration was measured using Qubit (Sigma). The lysis buffer was exchanged with UA buffer (8 M urea in 0.1 M Tris/ HCl, pH 8.5) by centrifugation using a Microcon filter unit (10 kDa cutoff (Millipore)). After incubation with IAA solution (0.05 M iodoacetamide in UA), excess reagent was removed by ultrafiltration. Samples were repeatedly washed with UB buffer (8 M urea in 0.1 M Tris/HCl, pH 8.0). The protein suspension was then digested overnight with endoproteinaseLys-C (1:30) at room temperature, and the resulting digest was diluted six 2360
dx.doi.org/10.1021/pr401163t | J. Proteome Res. 2014, 13, 2359−2369
Journal of Proteome Research
Article
the Swiss-prot Homo sapiens (human) protein sequence database containing 20306 sequences after taxonomy, using an in-house Mascot server (version 2.3.2, Matrix Science Ltd.). Trypsin was chosen as enzyme with a maximum of 2 missed cleavages. S-Carbamidomethyl cysteine was defined as a fixed modification. Partial modifications included were oxidation (M) and phosphorylation at serine, threonine, and tyrosine residues. The MS and MS/MS results were searched with a peptide ion mass tolerance of 10 ppm and a fragment ion mass tolerance of +0.8 Da. Percolator was used to calculate FDR (q-values) based on a decoy database search performed using a concatenated decoy database derived from the swiss-prot database. The gene names and UniProt and SwissProt IDs of the identified mitochondrial proteins were retrieved through Proteome Discoverer (ThermoFisher) together with Uniprot (http:/ www.uniprot.org/). Only peptides that were identified as peptide rank 1 and with a FDR
