Protein Phosphatase 1-α Regulates AS160 Ser588 and Thr642 Dephosphorylation in Skeletal Muscle
Akt substrate of 160 kDa (AS160) phosphorylation on Thr642 and Ser588 by Akt is essential for insulin’s full effect on glucose transport. However, protein phosphorylation is determined by the balance of actions by kinases and phosphatases, and the specific phosphatase(s) controlling AS160 dephosphorylation is (are) unknown. Accordingly, we assessed roles of highly expressed skeletal muscle serine/threonine phosphatases (PP1, PP2A, PP2B, and PP2C) on AS160 dephosphorylation. Preliminary screening of candidate phosphatases used an AS160 dephosphorylation assay. Lysates from insulin-stimulated skeletal muscle were treated with pharmacological phosphatase inhibitors and assessed for AS160 Ser588 and Thr642 dephosphorylation. AS160 dephosphorylation on both phosphorylation sites was unaltered by PP2B or PP2C inhibitors. Okadaic acid (low dose inhibits PP2A; high dose inhibits PP1) delayed AS160 Ser588 (both doses) and Thr642 (high dose only) dephosphorylation concomitant with greater Akt phosphorylation (both doses). AS160 was coimmunoprecipitated with PP1-α but not with PP1-β, PP1-γ1, or PP2A. Recombinant inhibitor-2 protein (a selective PP1 inhibitor) delayed AS160 dephosphorylation on both phosphorylation sites without altering Akt phosphorylation. Furthermore, knockdown of PP1-α but not PP1-β or PP1-γ1 by small interfering RNA caused greater AS160 Ser588 and Thr642 phosphorylation concomitant with unaltered Akt phosphorylation. Together, these results identified PP1-α as a regulator of AS160 Thr642 and Ser588 dephosphorylation in skeletal muscle.
Skeletal muscle accounts for the largest portion of insulin-mediated whole-body glucose disposal, and skeletal muscle insulin resistance is crucial for whole-body insulin resistance and type 2 diabetes (1). Muscle insulin resistance is secondary, in large part, to defective GLUT4 translocation and glucose transport (2). Insulin’s stimulation of glucose transport is triggered by a complex insulin-signaling pathway that begins with insulin’s binding to its receptor, leading to receptor autophosphorylation and activation of receptor tyrosine kinase (2). The insulin receptor kinase phosphorylates insulin receptor substrate (IRS) proteins on multiple tyrosine residues, resulting in IRS protein engagement with phosphatidylinositol (PI) 3-kinase (PI3K), that in turn, phosphorylates PI 4,5-bisphosphate to create 3,4,5-trisphosphate (PIP3). The serine/threonine kinase Akt is recruited to bind PIP3 and become activated secondary to phosphorylation on Thr308 via phosphoinositide-dependent kinase-1 (PDK1) and Ser473 via mTORC2. Akt phosphorylates many protein substrates, several of which have been implicated in insulin’s regulation of GLUT4 traffic to the cell surface membranes, including a Rab-GTPase activating protein known as Akt substrate of 160 kDa (AS160; also known as TBC1D4) (3–5). Akt can phosphorylate several residues on AS160. Mutation of serine or threonine to alanine to prevent phosphorylation of Ser588 or Thr642 resulted in attenuation of insulin-stimulated GLUT4 translocation, and mutation of several other Akt phosphomotifs did not produce any further effects on GLUT4 localization (6). Fully understanding the regulation of AS160 phosphorylation is essential given the crucial role that it plays in regulating insulin-stimulated glucose uptake by skeletal muscle.
The reversible serine/threonine phosphorylation of proteins is balanced by the opposing actions of kinases and phosphatases, but for most proteins, there has been an overwhelming bias to focus on serine/threonine kinases, with strikingly fewer studies assessing the role of serine/threonine phosphatases (7). Serine/threonine protein phosphatases regulate diverse aspects of growth, development, and metabolism, but relatively few protein serine/threonine phosphatases control the specific dephosphorylation of a much greater number of phosphoprotein substrates (8). With specific regard to AS160, many studies have analyzed the role of Akt in the insulin-stimulated phosphorylation of AS160 (9–13), but essentially nothing is known about the serine/threonine protein phosphatase(s) regulating AS160 dephosphorylation.
Protein phosphatase 1 (PP1), PP2A, PP2B, and PP2C are among the most abundant serine/threonine protein phosphatases expressed by skeletal muscle (14), and we hypothesized that AS160 dephosphorylation on Thr642 and Ser588 would be regulated by one or more of these enzymes. We evaluated the hypothesis using multiple approaches, including assessment of
the effects of several pharmacologic serine/threonine protein phosphatase inhibitors on AS160 Ser588 and Thr642 dephosphorylation;
the physical association of AS160 with serine/threonine protein phosphatases;
the influence of a selective inhibitor of PP1, known as inhibitor 2 (Inh-2) (15), on AS160 Ser588 and Thr642 phosphorylation; and
the consequences of knockdown of serine/threonine protein phosphatases by small interfering (si)RNA silencing on AS160 Ser588 and Thr642 phosphorylation.
These experiments identified PP1-α as a serine/threonine protein phosphatase that regulates AS160 Ser588 and Thr642 dephosphorylation in skeletal muscle.
Research Design and Methods
The reagents and apparatus for SDS-PAGE and nonfat dry milk (#170-6404XTU) were from Bio-Rad (Hercules, CA). MemCode Reversible Protein Stain (#24580) and bicinchoninic acid (#23227) protein assay kits and tissue protein extraction reagent (T-PER; #78510) were from Thermo Fisher (Waltham, MA). Luminata Forte Western HRP Substrate (#WBLUF0100) was from EMD Millipore (Billerica, MA). Sanguinarine chloride (#ALX-350-076) was from Enzo Life Sciences (Farmingdale, NY). FK-506 (#3631) was purchased from Tocris (Bristol, U.K.). Okadaic acid (OA; #459620) was purchased from Merck Millipore (Billerica, MA). Recombinant protein phosphatase Inh-2 (#P0755) was from New England Biolabs (Ipswich, MA). Anti-phosphorylated (p)AktThr308 (#9275), anti-pAktSer473 (#9271), anti-Akt (#4691), anti-pAS160Thr642 (#8881), anti-pAS160Ser588 (#8730), anti–PP1-α (#2582), anti-spinophilin (#14136), and anti-rabbit IgG horseradish peroxidase conjugate (#7074) were from Cell Signaling Technology (Danvers, MA). Anti-AS160 (#ABS54), anti–PP1-β (#07–1217), anti–PP1-γ1 (#07-1218), anti–α-tubulin (#04-1117), and normal rabbit IgG polyclonal antibody control (#12-370) were purchased from EMD Millipore. Anti-PP2Aα (#610556) was from BD Bioscience (San Jose, CA). PP1 Inh-2 antibody (#AF4719) was from R&D Biosystems (Minneapolis, MN). Anti–PP1-α (#sc-443), anti-GADD34 (#sc-8327), anti-goat IgG horseradish peroxidase conjugate (#sc-2020), and anti-mouse IgG horseradish peroxidase conjugate (#sc-2060), were from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-phostensin (#MB1057) was from BioWorld Technology, Inc. (St. Louis Park, MN). siRNA against rat PP1 catalytic subunits α (#L-100270-02-0020), β (#L-100263-02-0020), or γ1 (#L-096319-02-0020) and RNA interference-negative control (#D-001810-10-20) were purchased as SMARTpools from Dharmacon (Lafayette, CO). Protein G magnetic beads (#10004D), RNAiMAX transfection reagent (#13778-150), and DMEM (#11995) were from Life Technologies (Grand Island, NY).
Procedures for animal care were approved by the University of Michigan Committee on Use and Care of Animals. Male Wistar rats (aged 8–10 weeks) were from Harlan (Indianapolis, IN). Lean (Fa/Fa) and obese (fa/fa) male Zucker rats (aged 7–8 weeks) were from Charles River Laboratories (Wilmington, MA). Animals were provided with rodent chow (Lab Diet No. 5001; PMI Nutrition International, Brentwood, MO) ad libitum until 1700 h the night before the experiment, when food was removed. The next day at 1000 h to 1200 h, rats were anesthetized (intraperitoneal injection of sodium pentobarbital), and both epitrochlearis muscles were isolated and treated as described below.
Isolated epitrochlearis muscles were incubated in glass vials containing Krebs-Henseleit buffer, 0.1% BSA, 2 mmol/L sodium pyruvate, 6 mmol/L mannitol, without (basal) or with insulin at 0.6 nmol/L (for coimmunoprecipitation [Co-IP] assays described below) or 30 nmol/L (for dephosphorylation assays described below) for 30 min in a heated, shaking water bath at 35°C with continuous gassing (95% O2 and 5% CO2). Immediately after the incubation, muscles were blotted, rapidly trimmed of connective tissue, and freeze-clamped with liquid N2–cooled aluminum tongs. Frozen muscles were stored at −80°C until subsequent homogenization and analysis.
L6 Cell Culture and Treatment
L6 myoblasts were purchased from the American Type Culture Collection (Manassas, VA). L6 cells were cultured in DMEM, supplemented with 10% (vol/vol) FBS, 1% penicillin, and 100 μg/mL streptomycin in a humidified atmosphere with 5% CO2 at 37°C. Cells were washed twice with 1× PBS, and starved for 5 h in serum-free DMEM medium before 20 min of incubation without insulin (basal) or with insulin (100 nmol/L).
Muscle and Cell Lysate Preparation
Unless otherwise noted, epitrochlearis muscles were homogenized (TissueLyser II homogenizer; Qiagen Inc., Valencia, CA) using ice-cold T-PER buffer supplemented with 2.5 mmol/L sodium pyrophosphate, 1 mmol/L sodium vanadate, 1 mmol/L β-glycerophosphate, 1 μg/mL leupeptin, 1 μg/mL pepstatin, 1 μg/mL aprotinin, and 1 mmol/L phenylmethyl sulfonyl fluoride (PMSF). Homogenates were then transferred to microcentrifuge tubes and rotated (1 h, 4°C) before being centrifuged (15,000g, 20 min, 4°C). L6 cells were scraped in T-PER that was supplemented as described above, incubated (4°C, 20 min), and then centrifuged (15,000g, 20 min, 4°C). Total protein in supernatants from muscle or cell lysates was measured by the bicinchoninic acid method.
Epitrochlearis muscles and L6 cells used for Co-IP were processed as described above using Co-IP buffer (50 mmol/L HEPES [pH 7.5], 150 mmol/L NaCl, 1% IGEPAL, 10% glycerol) supplemented with 2.5 mmol/L sodium pyrophosphate, 1 mmol/L sodium vanadate, 1 mmol/L β-glycerophosphate, 1 μg/mL leupeptin, 1 μg/mL pepstatin, 1 μg/mL aprotinin, and 1 mmol/L PMSF.
Epitrochlearis muscles used for the dephosphorylation assay described below were homogenized for 2 min in cold T-PER buffer supplemented with 1 μg/mL leupeptin, 1 μg/mL pepstatin, 1 μg/mL aprotinin, and 1 mmol/L PMSF, but without phosphatase inhibitors (2.5 mmol/L sodium pyrophosphate, 1 mmol/L sodium vanadate, 1 mmol/L β-glycerophosphate). Lysates were centrifuged (15,000g, 2 min, 4°C), and the resultant supernatants were immediately used for the dephosphorylation assay described below.
AS160 Dephosphorylation Assay
Preliminary insights about the regulation of AS160 dephosphorylation were provided by testing the ability of several chemical phosphatase inhibitors, with differing specificity for inhibiting selected protein phosphatases, to delay the rate of AS160 dephosphorylation in lysates prepared from isolated epitrochlearis muscles. The AS160 dephosphorylation assay was a modification of a method previously used to identify the serine/threonine phosphatase that dephosphorylates calcium-binding protein 4 (16) The muscles used for this AS160 dephosphorylation assay were first incubated with 30 nmol/L insulin to ensure initially high levels of AS160 phosphorylation. Freeze-clamped muscles were rapidly homogenized in ice-cold buffer in the absence of protein phosphatase inhibitors. An initial aliquot (20 µL) was rapidly withdrawn from each muscle lysate and immediately mixed with an equal volume of 2× SDS loading buffer and heated (95°C, 6 min). This initial aliquot was denoted as the 0-min time point for the AS160 dephosphorylation assay. To the remaining lysate, a chemical protein phosphatase inhibitor (5 nmol/L OA for PP2A inhibition and 1,000 nmol/L OA for PP1 inhibition (17); 10 μmol/L sanguinarine for PP2C inhibition (18); or 100 ng/mL FK-506 for PP2B inhibition (19)) or an equal volume of vehicle (DMSO) was rapidly added. The lysates were then incubated at 37°C, with aliquots (20 µL) withdrawn at 30, 60, and 120 min. Upon withdrawal, each aliquot was rapidly mixed with an equal volume of 2× SDS loading buffer and heated to 95°C for 6 min. SDS-PAGE and immunoblotting were used to assess the phosphorylation levels of AS160Ser588 and AS160Thr642 in these aliquots. When Inh-2 protein was used, dephosphorylation assays were performed as described above for chemical inhibitors, except that epitrochlearis lysates were incubated with or without recombinant Inh-2 (rInh-2) protein (50 μg/mL lysate) rather than a chemical inhibitor. Dephosphorylation assays in lean Zucker (LZ) and obese Zucker (OZ) rats were performed as described above, except lysates were incubated without any phosphatase inhibitors.
IP was performed in lysates prepared from epitrochlearis muscles (300 μg total protein) and L6 cells (400 μg total protein) using AS160 antibody or normal rabbit IgG at 4°C overnight. The next morning, the protein-antibody complex was incubated with 50 µL magnetic protein G beads for 2 h at 4°C with gentle rotation. The antibody-protein-beads complex was washed three times with Co-IP buffer. The protein in the complex was then eluted with 30 μL of 2× SDS loading buffer and boiled before running on a polyacrylamide gel. Proteins were transferred to polyvinylidene fluoride membranes, and AS160-associated proteins were immunoblotted using antibodies against PP1-α, PP1-β, PP1-γ1, and PP2A. The association of AS160 with PP1-α was also assessed by IP using anti–PP1-α, followed by immunoblotting with anti-AS160. In addition, the association of AS160 with several PP1-α regulatory subunits (20–22) was evaluated by immunoprecipitation using anti-AS160, followed by immunoblotting with antibodies against spinophilin (also known as PPP1R9B), GADD34 (also known as PPP1R15A), phostensin (also known as PPP1R18), and Inh-2 (also known as PPP1R2).
L6 cells were transfected with 200 nmol/L control scrambled siRNA, siPP1-α, siPP1-β, or siPP1-γ1 using RNAiMAX transfection reagent from Invitrogen using the manufacturer's instructions, with minor modifications. Briefly, cells were seeded in 60-mm plates at 50% confluence in DMEM supplemented with 10% FBS. On day 2, myoblasts were transfected with 200 nmol/L of siRNA in reduced serum Opti-MEM media without antibiotics. The media changed 48 h later to DMEM supplemented with 10% FBS. At 72 h of transfection, the cells were treated without insulin (basal) or with 100 nmol/L insulin in serum-free medium for 20 min before being harvested. The extent of PP1-α, PP1-β, or PP1-γ1 knockdown was assessed by immunoblot.
Equal amounts of protein from each sample were loaded on a Tris glycine acrylamide gel and transferred to polyvinylidene fluoride membrane. The membranes were incubated with appropriate primary and secondary antibodies. Immunoreactive proteins were detected using Luminata Forte Western HRP Substrate and quantified by densitometry (Alpha Innotech, San Leandro, CA).
Statistical analyses were performed using Prism 4.0 software (GraphPad Software, La Jolla, CA). Data are expressed as means ± SEM. Differences between two groups were evaluated using a two-tailed t test. Differences between more than two groups were evaluated using one-way ANOVA. The source of significant variance was identified using Tukey post hoc analysis. A P value of ≤0.05 was considered statistically significant.
Neither FK-506 nor Sanguinarine Delay AS160 Dephosphorylation
Neither the PP2B inhibitor FK-506 nor the PP2C inhibitor sanguinarine differed from the vehicle for pAS160Thr642 (Fig. 1A and C) or pAS160Ser588 (Fig. 1B and C) in epitrochlearis muscle lysates. Similarly, neither inhibitor differed from vehicle for pAktThr308 (Fig. 1D and F) or pAktSer473 (Fig. 1E and F). These results provide no evidence that PP2B or PP2C regulate AS160 dephosphorylation.
Dose-Dependent Effects of OA on AS160 Dephosphorylation
There were no differences for the phosphorylation of AS160Thr642 of epitrochlearis muscle lysates treated with vehicle compared with 5 nmol/L OA, a dose sufficient to inhibit PP2A but not PP1 (23) at 30, 60, and 120 min. In contrast, for lysates treated with 1,000 nmol/L OA, a dose sufficient to inhibit PP1 (23), the phosphorylation of AS160Thr642 was significantly (P < 0.05) greater than the vehicle at 30, 60, and 120 min and significantly (P < 0.05) greater than the 5 nmol/L OA values at 30 and 60 min (Fig. 2A and C). For phosphorylation of AS160Ser588, the values with 5 nmol/L OA significantly (P < 0.05) exceeded the vehicle at 30 and 120 min, and the 1,000 nmol/L values were significantly (P < 0.05) greater than vehicle at 30, 60, and 120 min (Fig. 2B and C). For phosphorylation of AktThr308, there were no significant differences between the 5 nmol/L OA values and vehicle at 30, 60, and 120 min, but the 1,000 nmol/L values significantly (P < 0.05) exceeded both vehicle and 5 nmol/L OA values at 30, 60, and 120 min (Fig. 2D and F). For phosphorylation of AktSer473, the values with 5 nmol/L OA were significantly (P < 0.05) greater than vehicle at 30 min, and the 1,000 nmol/L OA values significantly (P < 0.05) exceeded vehicle and 5 nmol/L values at 30, 60, and 120 min (Fig. 2E and F). Taken together, these results suggest that AS160 dephosphorylation is influenced by OA in a dose-dependent and site-selective manner. Only the higher OA dose delayed dephosphorylation of AS160Thr642, which would be consistent with an effect related to inhibition of PP1. In contrast, both OA doses delayed dephosphorylation of AS160Ser588, suggesting possible roles of PP2A and/or PP1.
Selective AS160 Co-IP With PP1-α but Not PP1-β, PP1-γ1, or PP2A
Because the OA results were not definitive, we used additional approaches to probe the relationship of AS160 with PP1 and PP2A in skeletal muscle. Specific interaction between PP1-α and AS160 in epitrochlearis was indicated by the significantly (P < 0.05) greater amount of PP1-α that Co-IP with AS160 compared with normal IgG control regardless of insulin concentration (Fig. 3A and B). The specific association between AS160 and PP1-α was confirmed by IP using anti–PP1-α, followed by immunoblotting with anti-AS160, compared with normal IgG control (data not shown). In contrast, there was no evidence that AS160 had any specific association with PP-β1 (Fig. 3C and D), PP1-γ1 (Fig. 3E and F), or PP2A (Fig. 3G and H). Similar results were obtained for L6 cells in which specific interaction was detected for AS160 with PP1-α but not for AS160 with PP1-β, PP1-γ1, or PP2A (data not shown). The selective association of AS160 with PP1-α supports the idea that PP1-α may regulate AS160 dephosphorylation. The association of AS160 with several PP1-α regulatory subunits (spinophilin, GADD34, phostensin, and Inh-2) was evaluated in epitrochlearis lysates immunoprecipitated using anti-AS160, followed by immunoblotting with antibodies against the respective regulatory subunits. However, this analysis did not detect evidence of selective association of these proteins with AS160 (data not shown).
rInh-2 Protein Delays AS160 Dephosphorylation
Inh-2 was studied because it is a selective biological inhibitor of PP1 (15). Epitrochlearis lysates incubated with rInh-2 protein compared with control lysates had greater (P < 0.05) phosphorylation of AS160Thr642, at 15, 30, 45, and 60 min (Fig. 4A and C). Furthermore, rInh-2 treatment versus controls produced greater (P < 0.05) phosphorylation of AS160Ser588 at 15, 30, and 45 min (Fig. 4B and C). Greater AS160 phosphorylation was not accompanied by any significant effects of rInh-2 on the phosphorylation of AktThr308 (Fig. 4D and F) or AktSer473 (Fig. 4E and F). The delayed AS160 dephosphorylation on both Thr642 and Ser588 with unaltered Akt phosphorylation is consistent with PP1 being a modulator of AS160 dephosphorylation.
Silencing PP1-α but Not Other PP1 Isoforms Leads to Greater AS160 Phosphorylation
Because PP1-α was selectively associated with AS160 based on Co-IP, it was also critical to perform a functional assessment of the roles of PP1 isoforms on AS160’s phosphorylation status in intact muscle cells. L6 cells transfected with siPP1-α had a significant (P < 0.05) reduction in PP1-α protein abundance compared with cells transfected with scrambled control siRNA (Fig. 5A). The specificity of the PP1-α knockdown in the siPP1-α–transfected cells was evidenced by the lack of changes in PP1-β or PP1-γ1 protein levels (Fig. 5B). Lower PP1-α abundance led to significantly (P < 0.05) greater phosphorylation of AS160Thr642 (Fig. 5C) and AS160Ser588 (Fig. 5D) in the insulin-stimulated siPP1-α–transfected cells compared with insulin-stimulated control cells. Greater AS160 phosphorylation was not attributable to any effect of siPP1-α transfection on the phosphorylation of AktThr308 (Fig. 5F) or AktSer473 (Fig. 5G).
Transfection of L6 cells with siPP1-β produced a robust and significant (P < 0.05) reduction in PP1-β protein abundance compared with cells transfected with scrambled control siRNA (Fig. 6A). The specificity of the PP1-β knockdown was confirmed by the absence of changes in PP1-α or PP1-γ1 abundance (Fig. 6B). The lower PP1-β content had no detectable effects on the phosphorylation of AS160Thr642 (Fig. 6C), AS160Ser588 (Fig. 6D), AktThr308 (Fig. 6F), or AktSer473 (Fig. 6G) in siPP1-β–transfected cells compared with control cells.
Transfection of L6 cells with siPP1-γ1 caused a substantial and significant (P < 0.05) reduction in PP1-γ1 protein abundance compared with cells transfected with the scrambled control siRNA (Fig. 7A). The specificity of the PP1-γ1 knockdown was demonstrated by unchanged PP1-α or PP1-β levels (Fig. 7B). The lower PP1-γ1 abundance resulted in no changes in the phosphorylation of AS160Thr642 (Fig. 7C), AS160Ser588 (Fig. 7D), AktThr308 (Fig. 7F), or AktSer473 (Fig. 7G) in siPP1-γ1–transfected cells versus control cells.
AS160 Dephosphorylation in Muscles From LZ versus OZ Rats
The relative phosphorylation of AS160Thr642 was significantly (P < 0.05) lower for epitrochlearis muscle lysates from OZ versus LZ rats at 5 min (Fig. 8A and C). The relative phosphorylation of AS160Ser588 was significantly (P < 0.05) lower for OZ versus LZ at 30 and 40 min (Fig. 8B and C). In contrast to AS160 phosphorylation, there were no significant differences between LZ and OZ rats for phosphorylation of AktThr308 (Fig. 8D and F) or AktSer473 (Fig. 8E and F).
Epitrochlearis muscle lysates from LZ versus OZ did not differ for protein abundance of AS160 (P = 0.42), PP1-α (P = 0.78), or Inh-2 (P = 0.27) (data not shown). Neither did epitrochlearis muscle lysates from LZ versus OZ differ with regard to the amount of AS160-associated PP1-α (P = 0.40) determined by Co-IP (data not shown).
Protein phosphorylation status reflects the balance of phosphorylation by kinases and dephosphorylation by phosphatases. However, there is an extreme disparity in the level of knowledge about the roles of specific kinases compared with phosphatases in the regulation of protein phosphorylation. Although AS160 phosphorylation is a key determinant of insulin-stimulated glucose transport, prior research had not identified the serine/threonine protein phosphatases that regulate AS160 dephosphorylation. The current study tackled this important problem using a series of different experimental approaches leading to the discovery that PP1-α modulates AS160 dephosphorylation on Ser588 and Thr642 in skeletal muscle.
PP1, PP2A, PP2B, and PP2C are four of the most highly expressed serine/threonine phosphatases in skeletal muscle (14). We initially tested AS160 dephosphorylation in muscle using several pharmacologic inhibitors that are commonly used to block each of these phosphatases. This first screening step provided no evidence that PP2B or PP2C inhibitors delayed AS160 dephosphorylation on either AS160 phosphorylation site. OA was tested at two doses because it has differing potency for inhibiting PP2A (requiring a lower dose) compared with PP1 (requiring a higher dose) (23). The higher OA dose but not the lower OA dose delayed AS160 dephosphorylation on Thr642. This result is consistent with the possibility that PP1 regulated AS160 dephosphorylation on this site, but a caveat is that this OA dose also produced greater Akt phosphorylation. Both OA doses delayed AS160 dephosphorylation on Ser588 concomitant with increased Akt phosphorylation. Earlier research has indicated that OA can attenuate Akt dephosphorylation (24–26). The OA results did not conclusively isolate the possible roles of PP1 or PP2A for controlling AS160 dephosphorylation, so we subsequently used additional experimental approaches to gain more specific insights.
For our first alternative approach, we used Inh-2 to specifically assess the role of PP1 in regulating AS160 dephosphorylation. The results from the rInh-2 experiment were simpler to interpret because Inh-2 selectively inhibits PP1 without inhibiting PP2A (27) and because Inh-2 caused greater phosphorylation on both AS160 sites in the absence of altered Akt phosphorylation. These observations implicated PP1 in the regulation of AS160 dephosphorylation on both Ser588 and Thr642. However, because skeletal muscle expresses three PP1 isoforms (PP1-α, PP1-β, and PP1-γ1) and Inh-2 can bind and inhibit each of these PP1 isoforms (28), it was necessary to next address the possibility of PP1 isoform selectivity.
We used two distinct approaches to address the role of different PP1 isoforms. We first performed Co-IP analysis to evaluate the physical association between AS160 and candidate phosphatases (29). We found that AS160 was selectively associated with PP1-α and not PP1-β or PP1-γ1 in both rat skeletal muscle and L6 myocytes. There was also no evidence for specific interaction between AS160 and PP2A in rat skeletal muscle or L6 cells.
The Co-IP analysis demonstrated that PP1-α and AS160 are binding partners, but this approach cannot establish if there is a functional relationship between the two proteins. Accordingly, we next turned to siRNA silencing to test for PP1 isoform–specific effects. The results of this experiment were invaluable because they clearly revealed that a selective reduction in PP1-α protein abundance produced greater AS160 phosphorylation on Ser588 and Thr642. Importantly, Akt phosphorylation was unaltered by PP1-α knockdown, so the greater AS160 phosphorylation was not an indirect consequence of greater Akt phosphorylation. Furthermore, PP1-α knockdown was specific, because there were no changes in the abundance of the other PP1 isoforms. In contrast to the significant effects of PP1-α knockdown, neither PP1-β nor PP1-γ1 knockdown altered AS160 phosphorylation.
It seems possible that altered serine/threonine dephosphorylation of AS160 may contribute to attenuated AS160 phosphorylation that has been reported in insulin-resistant muscles. For example, OZ rats compared with LZ rats are characterized by reductions in insulin-stimulated AS160 phosphorylation and glucose uptake (30–33). The current study included the first evaluation of the potential role of accelerated AS160 dephosphorylation in insulin-resistant muscles. Results from the AS160 dephosphorylation assay revealed modestly faster dephosphorylation of AS160 on Ser588 and Thr642 in muscle lysates from insulin-resistant OZ rats versus LZ controls. These differences were not explained by disparities in the abundance of AS160, PP1-α, or Inh-2 protein or in the association between PP1-α and AS160. However, these results do not eliminate the possibility of a role for PP1-α in the dysregulation of AS160 phosphorylation in insulin-resistant muscle.
In conclusion, AS160 plays a pivotal role in insulin’s regulation of glucose transport in skeletal muscle. Many previous studies have focused on the role of Akt in the insulin-stimulated phosphorylation of AS160. In contrast, the current study was the first to focus on identifying the serine/threonine phosphatases that control the dephosphorylation of AS160. We discovered that PP1-α regulates AS160 dephosphorylation on Ser588 and Thr642, two key sites that control insulin-stimulated glucose transport. This knowledge represents an essential building block for fully understanding the processes that control this key insulin signaling protein that is a crucial regulator of insulin-stimulated glucose transport.
Acknowledgments. The authors acknowledge Naveen Sharma, PhD, currently of Central Michigan University, for his role in the initial development of the dephosphorylation assay.
Funding. This research was supported by a grant from the National Institutes of Health (R01-DK-071771).
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Author Contributions. P.S. performed the experiments, analyzed the data, designed the experiments, discussed the manuscript, developed the hypothesis, and wrote the manuscript. E.B.A. performed the experiments, analyzed the data, and discussed the manuscript. G.D.C. designed the experiments, coordinated and directed the project, developed the hypothesis, discussed the manuscript, and wrote the manuscript. G.D.C. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
- Received June 24, 2015.
- Accepted May 23, 2016.
- © 2016 by the American Diabetes Association.
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://diabetesjournals.org/site/license.