Temporal Dynamics of Tyrosine Phosphorylation in Insulin Signaling
http://www.100md.com
《糖尿病学杂志》
1 Biological Engineering Division, Massachusetts Institute of Technology, Cambridge, Massachusetts
2 Biochemistry Department, Dartmouth Medical School, Hanover, New Hampshire
APS, adaptor protein with pleckstrin homology and src homology 2 domains; ERK, extracellular signal–regulated kinase; Gab, growth factor receptor–bound protein 2–associated binding protein; IMAC, immobilized metal affinity chromatography; IRS, insulin receptor substrate; MAPK, mitogen-activated protein kinase; MS/MS, tandem mass spectrometry; PI3K, phosphatidylinositol 3-kinase; PTRF, polymerase I and transcript release factor; pTyr, phosphorylated tyrosine; SHC, Src homology 2 domain containing transforming protein C; SHP-2, protein tyrosine phosphatase, non-receptor type II
ABSTRACT
The insulin-signaling network regulates blood glucose levels, controls metabolism, and when dysregulated, may lead to the development of type 2 diabetes. Although the role of tyrosine phosphorylation in this network is clear, only a limited number of insulin-induced tyrosine phosphorylation sites have been identified. To address this issue and establish temporal response, we have, for the first time, carried out an extensive, quantitative, mass spectrometry-based analysis of tyrosine phosphorylation in response to insulin. The study was performed with 3T3-L1 adipocytes stimulated with insulin for 0, 5, 15, and 45 min. It has resulted in the identification and relative temporal quantification of 122 tyrosine phosphorylation sites on 89 proteins. Insulin treatment caused a change of at least 1.3-fold in tyrosine phosphorylation on 89 of these sites. Among the responsive sites, 20 were previously known to be tyrosine phosphorylated with insulin treatment, including sites on the insulin receptor and insulin receptor substrate-1. The remaining 69 responsive sites have not previously been shown to be altered by insulin treatment. They were on proteins with a wide variety of functions, including components of the trafficking machinery for the insulin-responsive glucose transporter GLUT4. These results show that insulin-elicited tyrosine phosphorylation is extensive and implicate a number of hitherto unrecognized proteins in insulin action.
Metabolic control is primarily regulated by the insulin-signaling network. In healthy individuals, insulin stimulates glucose uptake from the bloodstream into adipose tissue and skeletal muscle while inhibiting glucose production in the liver. Dysregulation of this network associated with insulin resistance causes an increase in blood glucose and lipid levels, often initially associated with an increase in insulin levels and eventually culminating in type 2 diabetes (1). Understanding the signaling network activated by insulin stimulation is crucial for identifying the causes and effects of network dysregulation and insulin resistance.
Insulin binds to the insulin receptor at the cell surface and activates its tyrosine kinase activity, leading to autophosphorylation and phosphorylation of several receptor substrates. Phosphorylation of selected tyrosine sites on receptor substrates is known to activate different pathways leading to increased glucose uptake, lipogenesis, and glycogen and protein synthesis, as well as to stimulation of cell growth (1,2). In addition to activation of these pathways by tyrosine phosphorylation, several mechanisms of downregulating the response to insulin stimulation have also been identified. For instance, serine phosphorylation on insulin receptor substrate (IRS)-1 induced by a variety of factors has been shown to interfere with the activating effects of tyrosine phosphorylation by decreasing binding to the insulin receptor or increasing degradation of IRS-1 (1,3,4). Ser/Thr phosphorylation of the insulin receptor has also been shown to decrease tyrosine kinase activity (1). Downregulation of the insulin receptor (1) and IRS-1 (3) are two additional mechanisms of mediating insulin resistance. Many of these factors are reflected in decreased amounts of the tyrosine-phosphorylated receptor and receptor substrates with concomitant reduction in downstream signaling. Even though tyrosine phosphorylation plays a key role in insulin signaling, rather limited knowledge of specific phosphorylation sites, mainly focusing on tyrosine phosphorylation on the insulin receptor and IRS-1, is available so far. Therefore, we expected that a more comprehensive analysis of tyrosine phosphorylation upon insulin stimulation would lead to further insights into the biology of the signaling network.
We have recently developed a mass spectrometric methodology for the identification and quantification of tyrosine phosphorylation sites on many proteins (5). Here we have applied this methodology to the analysis of insulin signaling in 3T3-L1 adipocytes stimulated with insulin for 0, 5, 15, or 45 min. Using this approach, we were able to identify and quantify the temporal dynamics of many previously described sites on the insulin receptor and several insulin receptor substrates, as well as many additional sites, both previously characterized and novel, on other proteins associated with insulin signaling. These include sites related to the mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) pathways. Moreover, this analysis also produced temporal phosphorylation profiles for a number of novel phosphorylation sites on proteins which have, so far, not been directly associated with insulin signaling, such as proteins in the machinery of GLUT4 trafficking. The results of this study show that the insulin-elicited increase in tyrosine phosphorylation is more widespread than previously known and identify many new sites to be explored for their specific roles in insulin action.
RESEARCH DESIGN AND METHODS
Cell culture, insulin stimulation, and cell lysis.
3T3-L1 fibroblasts from the American Type Culture Collection were carried as fibroblasts and differentiated into adipocytes, as described previously (6). Confluent 10-cm plates of adipocytes at day 7 after differentiation (1 x 107 cells per 10-cm plate) were washed with and incubated in serum-free Dulbecco's modified Eagle's medium with 1 mg/ml BSA for 16 h. Cells were then stimulated with 150 nmol/l insulin in this medium for 5, 15, and 45 min; nontreated cells were used as 0-min time point. For every time point, three plates were prepared. After rinsing with PBS, cells were lysed in 1.5 ml 8 mol/l urea containing 1 mmol/l Na3VO4 (7); the cell lysate was frozen in liquid nitrogen and stored at –70°C until further use.
Sample processing and peptide immunoprecipitation.
Protein concentration was determined by bicinchoninic acid protein concentration assay (Pierce). Proteins were reduced, alkylated, and digested with modified trypsin (Promega; enzyme-to-substrate ratio 1:50) (5). The digest was acidified to pH 2 with HCl, centrifuged, and the supernatant was filtered (Millex-HV filter, 0.45-μm pore size; Millipore), desalted, and fractionated on a C18 Sep-Pak Plus Cartridge (Waters). Peptides eluted with 25% acetonitrile in 0.1% acetic acid were lyophilized, labeled with iTRAQ (1/4 plate per condition), and combined, and phosphotyrosine (pTyr) peptide immunoprecipitation was performed with anti-pTyr antibodies, as previously described (5), in 30 mmol/l TrisCl, 30 mmol/l NaCl, pH 7.4 containing 0.4% Nonidet P40 Substitute (Fluka). For peptide immunoprecipitation, a mixture of P-Tyr-100 (18 μg; Cell Signaling no. 9411) and PT-66 (12 μg; Sigma P3300), previously coupled to Protein G Plus-Agarose beads (20 μl; Calbiochem), was used, with the peptides derived from one 10-cm plate in 0.45 ml. Bound peptides were eluted from the antibody with 70 μl glycine 100 mmol/l, pH 2.1.
Chromatography and liquid chromatography tandem mass spectrometry.
Since peptide immunoprecipitation with pan-specific anti-phosphotyrosine antibodies is subject to nonspecific binding (specifically, those peptides containing aromatic amino acids [e.g., tyrosine, phenylalanine, and tryptophan] appear to be preferentially enriched), further enrichment for phosphopeptides was performed using immobilized metal affinity chromatography (IMAC), as previously described (5). This tandem affinity strategy virtually eliminated nonspecifically retained peptides, such that almost all peptides in the final analysis contained phosphorylated tyrosine. Phosphopeptides were eluted from the IMAC column to a capillary precolumn (100 μm i.d., packed with 10 μm ODS-A [Kanematsu]), which was then connected to a capillary analytical column (50 μm i.d., packed with 10 cm of 5 μm ODS-AQ [YMC-Waters]) with an integrated, laser-pulled (Model P-2000; Sutter Instrument) electrospray ionization emitter tip (2-μm diameter) (8). Peptides were eluted (flow rate 20 nl/min) from the liquid chromatography column to the quadrupole time-of-flight mass spectrometer (QSTAR XL; Applied Biosystems) with the following gradient: 0 min: 0% B; 10 min: 13% B; 105 min: 42% B; 115 min: 60% B; 122 min: 100% B (solvent A = 0.2 mol/l acetic acid and solvent B = 70% acetonitrile, 0.2 mol/l acetic acid). Data were acquired in information-dependent acquisition mode, in which a full scan mass spectrum (2.5 s) was followed by tandem mass spectrometry (MS/MS) of the four most abundant ions (4 s each) of charge state 2–5 using mass exclusion time of 25 s.
Phosphopeptide sequencing and quantitative analysis.
MS/MS spectra were extracted and searched against a rodent (mouse and rat) protein database (NCBI) using ProQuant (Applied Biosystems) as described previously (5). Mass tolerance was set to 2.2 atomic mass units for precursor ions and 0.15 atomic mass units for fragment ions. Phosphotyrosine-containing peptides were manually validated and quantified. Peak areas for each of the four signature peaks (m/z: 114, 115, 116, and 117) were obtained and corrected according to the manufacturer's instructions to account for isotopic overlap. Only spectra with signature peaks below 1,500 counts were considered for quantification. To compensate for small differences in the sample amounts at each time point, the results were normalized to those for nonphosphorylated peptides of 10 abundant proteins present in the samples (online appendix Table 1 [available at http://diabetes.diabetesjournals.org]). Finally, all data were normalized by the 5-min sample.
The complete analysis of tyrosine phosphorylation was performed as described above three separate times, starting with lysates from separate 10-cm plates.
RESULTS
Overview of insulin-elicited tyrosine phosphorylation.
To quantify temporal dynamics of tyrosine phosphorylation in the insulin-signaling network, we have immunoprecipitated stable isotope-coded, tyrosine-phosphorylated peptides from 3T3-L1 adipocytes stimulated with insulin for either 0, 5, 15, or 45 min (Fig. 1). IMAC–liquid chromatography MS/MS analysis of the immunoprecipitated samples generated quantitative, temporal phosphorylation profiles for 126 peptides from 89 proteins. More than 80% of the peptides were quantified from at least two of the three biological replicates with an average SD of 10% for the three analyses (online appendix, Table 2).
All the sites identified and quantified in this study are listed in Table 1 according to their changes in phosphorylation after 5-min insulin stimulation relative to control (no insulin stimulation). Table 2 of the online appendix provides a complete presentation of the data. In the description of the results given below, if the supporting data are not presented in a table or figure, they can be found in Table 1 of the text and/or Table 2 of the online appendix.
An overview of the results is as follows. Of the 122 sites in 89 proteins that were analyzed, the phosphorylation level of 86 sites in 68 proteins increased by a factor of 1.3 or more in response to insulin, while three sites on three proteins decreased by a factor of 1.3 or more. The remaining 33 sites showed <1.3-fold change in phosphorylation after 5-min insulin treatment. Among the 89 sites that were altered in their extent of phosphorylation >1.3-fold in response to insulin, 38 sites have not previously been identified as sites of tyrosine phosphorylation in any context. Moreover, among the 51 responsive sites previously identified in any context, there were 20 sites previously known to undergo a change in tyrosine phosphorylation in response to insulin or the closely related IGF-I. The remaining 31 sites of tyrosine phosphorylation have been identified in other contexts but were not previously known to be affected by insulin treatment. Thus, overall we have identified 69 sites with a response of at least 1.3-fold that are either entirely novel or novel in the context of insulin action.
Especially notable in the dataset are 12 peptides that increased in phosphorylation by >10-fold following 5 min of insulin stimulation (Table 2). These included all the peptides containing sites identified on the insulin receptor itself. In addition, several sites on proteins related to the MAPK pathway were found in this group: Y1171 on IRS-1 and Y660 on growth factor receptor–bound protein 2–associated binding protein (Gab) 1; two sites with very similar sequences (pYI/LDLDL) that bind protein tyrosine phosphatase, non-receptor type II (SHP-2) and thereby participate in activation of the MAPK pathway (9,10); Y53 on Sprouty4, a site known to have inhibitory effect on MAPK activation (11); and the doubly phosphorylated (T202 and Y204, T185 and Y187) active forms of extracellular signal–regulated kinase (ERK) 1 and 2 (12). Three other phosphorylation sites were also found to have >10-fold increase in phosphorylation: Y618 on adaptor protein with pleckstrin homology and src homology 2 domains (APS), an adaptor protein linking the insulin receptor to Cbl binding (13); Y521 on Munc18c, a novel site on this protein, which is involved in the fusion of GLUT4 vesicles with the plasma membrane (14); and Y1640 on Cdc42bpb, a novel site on this serine kinase, which may act as a downstream effector of Cdc42 in cytoskeleton reorganization (15). In the following sections, we describe in more detail various sets of tyrosine phosphorylation sites.
The head of the pathway: insulin receptor and substrates of the insulin receptor.
Following ligand binding, the insulin receptor autophosphorylates on selected tyrosine residues, increasing kinase activity and recruiting adaptor proteins and substrates. We detected the singly and doubly phosphorylated forms of the catalytic loop of the kinase domain (including Y1175, 1179, and 1180) with strong increase in phosphorylation up to 5 min and slight decrease to 45 min. In vitro, the triply phosphorylated form has been reported to lead to full activation of the phosphotransferase activity; however, in vivo, the doubly phosphorylated form was found to be the major form (16). The COOH-terminal tyrosines 1345 and 1351 were also identified to be phosphorylated with a time profile comparable to the sites in the catalytic loop. Those two sites have been reported to be involved in the regulation of the phosphotransferase activity of the insulin receptor (17).
IRS-1, IRS-2, Src homology two domain containing transforming protein C (SHC), Gab1, and APS are insulin receptor substrate/scaffolding proteins, the tyrosine phosphorylation of which connects the activation of the insulin receptor to specific pathways (1,2,10,18,19). Mouse knockouts of IRS-1 and IRS-2 have marked phenotypes that show these two IRSs play particularly prominent roles in insulin signaling (20). As expected, insulin caused a marked increase in the phosphorylation of one or more tyrosines on each of these substrate/scaffolding proteins. In the case of IRS-1, we were able to monitor the phosphorylation of 4 (Y460, Y935, Y983, Y1171) of the 10 tyrosine phosphorylation sites previously reported for mouse, rat, or human IRS-1 (PhosphoSite). Three of the sites were in the pYXXM motif, which is known to bind to the SH2 domains of the regulatory subunit of PI3K, and, as noted above, the fourth (Y1171) is a binding site for SHP-2 (10). They all showed maximal phosphorylation after 5-min insulin stimulation followed by a slight decrease with longer insulin stimulation. The peptide containing phosphorylation of Y460 demonstrated a greater decrease (to about 50% of maximum level), which was most likely due to the appearance of the doubly phosphorylated form (T448/Y460) with maximum intensity after 15-min stimulation. The T448 site has not previously been reported and its function is not known. However, as noted earlier, there are many phosphorylation sites on serine residues of IRS-1 that have been shown to alter its extent of tyrosine phosphorylation (3,4).
In the case of IRS-2, we identified and quantified six tyrosine phosphorylation sites: Y594, Y628, Y649, Y671, Y734, Y758, and Y814. Remarkably, although the sites of tyrosine phosphorylation on IRS-2 have been inferred by comparison to those on IRS-1 (21), to our knowledge, they have not been previously determined. All but one of these sites is in the motif pYXXM, which binds to SH2 domains of the regulatory subunit of PI3K. For all of them, with the exception of Y671, phosphorylation increased more than twofold from 0 to 5 min followed by mostly constant levels up to 45 min. Y671 showed only slight increase from 0 to 5 min, and Y734 decreased slightly from 5 to 45 min. However, the peptide containing Y734 was also observed in the doubly phosphorylated form with a serine phosphorylation at position 727 or 728; the doubly phosphorylated peptide slightly increased (20%) from 5 to 45 min, potentially offsetting the decrease of the singly phosphorylated form. As described above, two other substrate/scaffolding proteins, Gab1 and APS, showed a marked increase in tyrosine phosphorylation. In addition, there was an eightfold increase in the tyrosine phosphorylation of SHC, which generates a motif that binds to the adaptor protein Grb2 and contributes to activation of the MAPK pathway (22).
Proteins associated with the insulin receptor and/or the IRSs.
In addition to the substrates of the insulin receptor described above, insulin treatment caused increases in tyrosine phosphorylation on a number of other proteins known to be associated with the insulin receptor and/or its scaffolding substrates and to participate in insulin signaling. The tyrosine phosphatases, hemopoietic cell phosphatase, which can associate directly with the insulin receptor, and SHP-2, which associates with the IRSs and Gab1, underwent tyrosine phosphorylation on previously identified sites (23,24). The adaptor protein, Crk, and, to a small extent, the related adaptor protein CrkL, were tyrosine phosphorylated on previously undescribed sites. Crk and CrkL are known to associate with both the insulin receptor and the IRSs (25,26). The -type 85 kDa regulatory subunit of PI3K exhibited 1.5-fold increase in tyrosine phosphorylation of Y467 and Y580. Y580, but not Y467 (note that this peptide could be derived from either p85- or p55-), has previously been reported to be phosphorylated by the insulin receptor (27). PI3K binds through the SH2 domains on its regulatory subunit to the IRSs and is thereby activated to produce the key signaling lipid, PtdIns 3,4,5-trisphosphate, which in turn activates the kinase Akt (1,2). SHIP2, a phosphatase for the five phosphoryl group on the signaling lipid PtdIns 3,4,5 trisphosphate, showed an eightfold increase in phosphorylation on Tyr 887, another site not previously identified as responsive to insulin. SHIP2 binds to SHC, and may participate in shutting down insulin signaling through the PI3K/Akt pathway (28). Nck 1 and 2, closely related adaptor proteins that bind to IRS-1 and may serve as a link to the cytoskeleton (29,30), increased in tyrosine phosphorylation on a single site 1.8-fold. This site on Nck has previously been identified to undergo phosphorylation in response to epidermal growth factor but not insulin (5). Fer, a cytoplasmic tyrosine kinase that associates with IRS-1 in insulin-treated adipocytes (31), exhibited a 1.7-fold increase in phosphorylation on Y402. Finally, in contrast to the other proteins in this section, the Src family tyrosine kinase Fyn, which associates with tyrosine phosphorylated IRS-1 (32), underwent a 40% decrease in phosphorylation on Y417 over the 45 min of insulin exposure. This site is due to autophosphorylation and increases kinase activity (33), so dephosphorylation may be associated with reduced Fyn kinase activity. (The peptide containing this site is also found in three other Src tyrosine kinase family members, Yes, Src, and Lck. We have assigned this site to Fyn based on previous reports linking Fyn to IRS-1, but it is possible that the quantification of this site may reflect the sum total of these Src family members.)
Other known insulin-elicited phosphotyrosine proteins in adipocytes.
Among the pTyr proteins, there is a group that has previously been identified as undergoing insulin-stimulated tyrosine phosphorylation in adipocytes but whose roles in signaling, if any, are less clear. This group consists of caveolin 1 and 2 (34,35), polymerase I and transcript release factor (PTRF) (36), Syncrip (37), and fatty acid binding protein 4 (also known as aP2) (38). Caveolin 1 and 2 and PTRF are major protein components of the caveolar regions in the plasma membrane, which also contain insulin receptors (34,39). In this regard, we also found a novel insulin-elicited phosphorylation on another caveolar protein, known as sdr. Sdr has previously been found to be the protein in caveolae that binds protein kinase C (40). The sites of insulin-stimulated phosphorylation on caveolin 1/2 and fatty acid binding protein 4, but not those on PTRF and Syncrip, have been previously identified.
Proteins not previously associated with the insulin-signaling network.
In addition to the many proteins and phosphorylation sites which have been associated with insulin signaling, a substantial portion of the phosphorylation sites are on proteins that have not previously been reported to be directly involved in insulin signaling. In this section, we describe some, but not all, of these.
A major effect of insulin in adipocytes is the stimulation of glucose transport. The basis for this effect is the rapid introduction of additional glucose transporters of the GLUT4 type into the plasma membrane in response to insulin. The latter is achieved via the insulin-triggered movement of specialized intracellular vesicles containing GLUT4 to the plasma membrane and their fusion therewith (41). This process is referred to as GLUT4 translocation. Thus, it is of considerable interest that we found four proteins previously implicated in GLUT4 translocation, Syntaxin4, Munc18c, EH domain-containing protein 2, and Annexin II, to be tyrosine phosphorylated in response to insulin (Table 1 and Fig. 2). Syntaxin4 is the plasma membrane snare protein that associates with the snare VAMP2 on GLUT4 vesicles (41). Munc18c associates with syntaxin4 and thereby may inhibit the binding of GLUT4 vesicles to syntaxin4 in the absence of insulin (14). EH domain-containing protein 2 has been found to associate with GLUT4 (42), and it participates in GLUT4 endocytosis (43). Annexin II appears to be involved in GLUT4 translocation in a way that is not yet defined (44). Current evidence indicates that the signaling pathway to GLUT4 translocation proceeds from the insulin receptor through PI3K and the serine kinase Akt, which is not known to stimulate any tyrosine kinase (45). Thus, it is unclear whether any of these tyrosine phosphorylation sites are part of the signaling network leading to GLUT4 translocation, but because they are on proteins of the trafficking machinery, they deserve consideration.
Two transporters, a K-Cl cotransporter (Slc12a4) and an amino acid transporter (Slc38a2, SAT2), were tyrosine phosphorylated in response to insulin. Insulin has been reported to stimulate both of these transport systems. However, in the case of the K-Cl transporter, the basis for the stimulation appears to be an increase in transporter mRNA (46), and in the case of the amino acid transporter, translocation from an internal source, similar to that for GLUT4, has been found to underlie the stimulation (47). It may be that tyrosine phosphorylation also contributes to the stimulation.
A group of four of the stimulated pTyr proteins may participate in adhesions between adipocytes. These are three membrane-associated guanylate kinases (MPP1, MPP7, and SAP102), members of a family of proteins known to be involved in cell-to-cell adhesion (48), and the adipocyte adhesion protein, a transmembrane protein that has been shown to participate in cell aggregation (49). Thus, it is possible that insulin treatment may alter the adipocyte/adipocyte interactions.
Temporal dynamics of phosphorylation in response to insulin stimulation.
Sites that increase in phosphorylation throughout the time course (Fig. 4B) also cluster into proteins with similarities. There are two likely RNA-binding proteins, Syncrip and Hdlbp, and there are three membrane-associated guanylate kinases, Mpp 1 and 7 and SAP102. In addition, annexin II, which is in this group, is similar to the membrane-associated guanylate kinases in that all are peripheral membrane proteins associated with the plasma membrane (48,52).
It is interesting that sites with similar temporal profiles appear to group together in modules within the signaling network. We have made the same observation in the epidermal growth factor receptor signaling network (5). Such temporal analysis has the potential to reveal functions for previously uncharacterized sites and proteins.
DISCUSSION
This study is, to date, the most comprehensive analysis of the effect of insulin stimulation on site-specific protein tyrosine phosphorylation. We chose to use 3T3-L1 adipocytes for the analysis, since this cell type has been extensively studied, is highly insulin responsive, and is a model for the animal fat cell. Cells were continuously exposed to insulin for 0, 5, 15, or 45 min. Phosphotyrosine peptide immunoprecipitation combined with iTRAQ stable isotope labeling and IMAC–liquid chromatography MS/MS produced quantitative temporal phosphorylation profiles for a large proportion of the sites and proteins in the insulin-signaling network. For instance, we detected phosphorylation on many of the previously known sites of insulin-elicited tyrosine phosphorylation on known proteins. Identification of these sites provides validation of the method. It is worth noting that not every previously known site was detected in this analysis, due to a variety of reasons, including incompatibility with IMAC–liquid chromatography MS/MS analysis or low signal level for a given peptide. In several cases (e.g., c-Cbl Y369, Gab-1 Y407), sites were identified but could not be quantified due to low signal level for iTRAQ marker ions or interfering, coeluting peptides. In addition to previously known sites, we found 66 sites with at least 1.3-fold increase in phosphorylation upon 5-min insulin stimulation that had not previously been known to be insulin responsive. Thus, insulin-elicited tyrosine phosphorylation is much more extensive than previously known.
The significance for insulin action of tyrosine phosphorylation on each of the newly discovered sites remains to be determined. The present study offers many encouraging leads, such as the tyrosine phosphorylation of a group of proteins that are components in the trafficking machinery for GLUT4. The elucidation of the function of each tyrosine phosphorylation will require an in-depth investigation of the particular site on the particular protein. A general approach to this problem is to knock down the protein in question, replace it with mutant protein containing nonphosphorylatable phenylalanine in place of tyrosine, and examine the effect of the replacement on insulin action. Clearly, further characterization of these novel sites and proteins will result in a significant expansion of our knowledge of the insulin-signaling network.
In the future, application of this approach to the investigation of insulin signaling in various tissues of normal and diabetic mice, such as adipose tissue, muscle, and liver, will be a powerful approach to identify tissue-specific sites of tyrosine phosphorylation and the specific alterations in tyrosine phosphorylation in the diabetic state.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health Grant DK42816.
We thank other members of the White lab and Sampsa Hautaniemi in the Lauffenburger lab at MIT for their assistance and helpful discussions.
FOOTNOTES
Additional information for this article can be found in an online appendix at http://diabetes.diabetesjournals.org.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Pirola L, Johnston AM, Van Obberghen E: Modulation of insulin action. Diabetologia 47:170–184, 2004
Saltiel AR, Kahn CR: Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414:799–806, 2001
Gual P, Le Marchand-Brustel Y, Tanti JF: Positive and negative regulation of insulin signaling through IRS-1 phosphorylation. Biochimie 87:99–109, 2005
Zick Y: Uncoupling insulin signalling by serine/threonine phosphorylation: a molecular basis for insulin resistance. Biochem Soc Trans 32:812–816, 2004
Zhang Y, Wolf-Yadlin A, Ross PL, Pappin DJ, Rush J, Lauffenburger DA, White FM: Time-resolved mass spectrometry of tyrosine phosphorylation sites in the epidermal growth factor receptor signaling network reveals dynamic modules. Mol Cell Proteomics 4:1240–1250, 2005
Frost SC, Lane MD: Evidence for the involvement of vicinal sulfhydryl groups in insulin-activated hexose transport by 3T3–L1 adipocytes. J Biol Chem 260:2646–2652, 1985
Gordon JA: Use of vanadate as protein-phosphotyrosine phosphatase inhibitor. Methods Enzymol 201:477–482, 1991
Martin SE, Shabanowitz J, Hunt DF, Marto JA: Subfemtomole MS and MS/MS peptide sequence analysis using nano-HPLC micro-ESI fourier transform ion cyclotron resonance mass spectrometry. Anal Chem 72:4266–4274, 2000
Cunnick JM, Mei L, Doupnik CA, Wu J: Phosphotyrosines 627 and 659 of Gab1 constitute a bisphosphoryl tyrosine-based activation motif (BTAM) conferring binding and activation of SHP2. J Biol Chem 276:24380–24387, 2001
White MF: IRS proteins and the common path to diabetes. Am J Physiol Endocrinol Metab 283:E413–E422, 2002
Kim HJ, Bar-Sagi D: Modulation of signalling by Sprouty: a developing story. Nat Rev Mol Cell Biol 5:441–450, 2004
Her JH, Lakhani S, Zu K, Vila J, Dent P, Sturgill TW, Weber MJ: Dual phosphorylation and autophosphorylation in mitogen-activated protein (MAP) kinase activation. Biochem J 296:25–31, 1993
Ahn MY, Katsanakis KD, Bheda F, Pillay TS: Primary and essential role of the adaptor protein APS for recruitment of both c-Cbl and its associated protein CAP in insulin signaling. J Biol Chem 279:21526–21532, 2004
Kanda H, Tamori Y, Shinoda H, Yoshikawa M, Sakaue M, Udagawa J, Otani H, Tashiro F, Miyazaki J, Kasuga M: Adipocytes from Munc18c-null mice show increased sensitivity to insulin-stimulated GLUT4 externalization. J Clin Invest 115:291–301, 2005
Leung T, Chen XQ, Tan I, Manser E, Lim L: Myotonic dystrophy kinase-related Cdc42-binding kinase acts as a Cdc42 effector in promoting cytoskeletal reorganization. Mol Cell Biol 18:130–140, 1998
White MF, Shoelson SE, Keutmann H, Kahn CR: A cascade of tyrosine autophosphorylation in the beta-subunit activates the phosphotransferase of the insulin receptor. J Biol Chem 263:2969–2980, 1988
Tennagels N, Bergschneider E, Al-Hasani H, Klein HW: Autophosphorylation of the two C-terminal tyrosine residues Tyr1316 and Tyr1322 modulates the activity of the insulin receptor kinase in vitro. FEBS Lett 479:67–71, 2000
Holgado-Madruga M, Emlet DR, Moscatello DK, Godwin AK, Wong AJ: A Grb2-associated docking protein in EGF- and insulin-receptor signalling. Nature 379:560–564, 1996
Ahmed Z, Pillay TS: Functional effects of APS and SH2-B on insulin receptor signalling. Biochem Soc Trans 29:529–534, 2001
Biddinger SB, Kahn CR: From mice to men: insights into the insulin resistance syndromes. Annu Rev Physiol 68:123–158 2005
Sun XJ, Wang LM, Zhang Y, Yenush L, Myers MG Jr, Glasheen E, Lane WS, Pierce JH, White MF: Role of IRS-2 in insulin and cytokine signalling. Nature 377:173–177, 1995
Sasaoka T, Kobayashi M: The functional significance of Shc in insulin signaling as a substrate of the insulin receptor. Endocr J 47:373–381, 2000
Uchida T, Matozaki T, Noguchi T, Yamao T, Horita K, Suzuki T, Fujioka Y, Sakamoto C, Kasuga M: Insulin stimulates the phosphorylation of Tyr538 and the catalytic activity of PTP1C, a protein tyrosine phosphatase with Src homology-2 domains. J Biol Chem 269:12220–12228, 1994
Stein-Gerlach M, Kharitonenkov A, Vogel W, Ali S, Ullrich A: Protein-tyrosine phosphatase 1D modulates its own state of tyrosine phosphorylation. J Biol Chem 270:24635–24637, 1995
Feller SM: Crk family adaptors-signalling complex formation and biological roles. Oncogene 20:6348–6371, 2001
Klammt J, Barnikol-Oettler A, Kiess W: Mutational analysis of the interaction between insulin receptor and IGF-I receptor with c-Crk and Crk-L in a yeast two-hybrid system. Biochem Biophys Res Commun 325:183–190, 2004
Hayashi H, Nishioka Y, Kamohara S, Kanai F, Ishii K, Fukui Y, Shibasaki F, Takenawa T, Kido H, Katsunuma N, et al.: The alpha-type 85-kDa subunit of phosphatidylinositol 3-kinase is phosphorylated at tyrosines 368, 580, and 607 by the insulin receptor. J Biol Chem 268:7107–7117, 1993
Dyson JM, Kong AM, Wiradjaja F, Astle MV, Gurung R, Mitchell CA: The SH2 domain containing inositol polyphosphate 5-phosphatase-2: SHIP2. Int J Biochem Cell Biol 37:2260–2265, 2005
Tu Y, Liang L, Frank SJ, Wu C: Src homology 3 domain-dependent interaction of Nck-2 with insulin receptor substrate-1. Biochem J 354:315–322, 2001
Rivera GM, Briceno CA, Takeshima F, Snapper SB, Mayer BJ: Inducible clustering of membrane-targeted SH3 domains of the adaptor protein Nck triggers localized actin polymerization. Curr Biol 14:11–22, 2004
Iwanishi M, Czech MP, Cherniack AD: The protein-tyrosine kinase fer associates with signaling complexes containing insulin receptor substrate-1 and phosphatidylinositol 3-kinase. J Biol Chem 275:38995–39000, 2000
Sun XJ, Pons S, Asano T, Myers MG Jr, Glasheen E, White MF: The Fyn tyrosine kinase binds Irs-1 and forms a distinct signaling complex during insulin stimulation. J Biol Chem 271:10583–10587, 1996
Roskoski R Jr: Src protein-tyrosine kinase structure and regulation. Biochem Biophys Res Commun 324:1155–1164, 2004
Kimura A, Mora S, Shigematsu S, Pessin JE, Saltiel AR: The insulin receptor catalyzes the tyrosine phosphorylation of caveolin-1. J Biol Chem 277:30153–30158, 2002
Lee H, Park DS, Wang XB, Scherer PE, Schwartz PE, Lisanti MP: Src-induced phosphorylation of caveolin-2 on tyrosine 19: phospho-caveolin-2 (Tyr(P)19) is localized near focal adhesions, remains associated with lipid rafts/caveolae, but no longer forms a high molecular mass hetero-oligomer with caveolin-1. J Biol Chem 277:34556–34567, 2002
Ibarrola N, Molina H, Iwahori A, Pandey A: A novel proteomic approach for specific identification of tyrosine kinase substrates using [13C]tyrosine. J Biol Chem 279:15805–15813, 2004
Hresko RC, Mueckler M: Identification of pp68 as the tyrosine-phosphorylated form of SYNCRIP/NSAP1: a cytoplasmic RNA-binding protein. J Biol Chem 277:25233–25238, 2002
Hresko RC, Hoffman RD, Flores-Riveros JR, Lane MD: Insulin receptor tyrosine kinase-catalyzed phosphorylation of 422(aP2) protein: substrate activation by long-chain fatty acid. J Biol Chem 265:21075–21085, 1990
Aboulaich N, Vainonen JP, Stralfors P, Vener AV: Vectorial proteomics reveal targeting, phosphorylation and specific fragmentation of polymerase I and transcript release factor (PTRF) at the surface of caveolae in human adipocytes. Biochem J 383:237–248, 2004
Mineo C, Ying YS, Chapline C, Jaken S, Anderson RG: Targeting of protein kinase Calpha to caveolae. J Cell Biol 141:601–610, 1998
Watson RT, Kanzaki M, Pessin JE: Regulated membrane trafficking of the insulin-responsive glucose transporter 4 in adipocytes. Endocr Rev 25:177–204, 2004
Park SY, Ha BG, Choi GH, Ryu J, Kim B, Jung CY, Lee W: EHD2 interactswith the insulin-responsive glucose transporter (GLUT4) in rat adipocytes and may participate in insulin-induced GLUT4 recruitment. Biochemistry 43:7552–7562, 2004
Guilherme A, Soriano NA, Furcinitti PS, Czech MP: Role of EHD1 and EHBP1 in perinuclear sorting and insulin-regulated GLUT4 recycling in 3T3–L1 adipocytes. J Biol Chem 279:40062–40075, 2004
Huang J, Hsia SH, Imamura T, Usui I, Olefsky JM: Annexin II is a thiazolidinedione-responsive gene involved in insulin-induced glucose transporter isoform 4 translocation in 3T3–L1 adipocytes. Endocrinology 145:1579–1586, 2004
Welsh GI, Hers I, Berwick DC, Dell G, Wherlock M, Birkin R, Leney S, Tavare JM: Role of protein kinase B in insulin-regulated glucose uptake. Biochem Soc Trans 33:346–349, 2005
Shen MR, Lin AC, Hsu YM, Chang TJ, Tang MJ, Alper SL, Ellory JC, Chou CY: Insulin-like growth factor 1 stimulates KCl cotransport, which is necessary for invasion and proliferation of cervical cancer and ovarian cancer cells. J Biol Chem 279:40017–40025, 2004
Hyde R, Peyrollier K, Hundal HS: Insulin promotes the cell surface recruitment of the SAT2/ATA2 system A amino acid transporter from an endosomal compartment in skeletal muscle cells. J Biol Chem 277:13628–13634, 2002
Funke L, Dakoji S, Bredt DS: Membrane-associated guanylate kinases regulate adhesion and plasticity at cell junctions. Annu Rev Biochem 74:219–245, 2005
Eguchi J, Wada J, Hida K, Zhang H, Matsuoka T, Baba M, Hashimoto I, Shikata K, Ogawa N, Makino H: Identification of adipocyte adhesion molecule (ACAM), a novel CTX gene family, implicated in adipocyte maturation and development of obesity. Biochem J 387:343–353, 2005
Kayali AG, Austin DA, Webster NJ: Stimulation of MAPK cascades by insulin and osmotic shock: lack of an involvement of p38 mitogen-activated protein kinase in glucose transport in 3T3–L1 adipocytes. Diabetes 49:1783–1793, 2000
Waters SB, Yamauchi K, Pessin JE: Insulin-stimulated disassociation of the SOS-Grb2 complex. Mol Cell Biol 15:2791–2799, 1995
Gerke V, Moss SE: Annexins and membrane dynamics. Biochim Biophys Acta 1357:129–154, 1997(Katrin Schmelzle, Susan K)
2 Biochemistry Department, Dartmouth Medical School, Hanover, New Hampshire
APS, adaptor protein with pleckstrin homology and src homology 2 domains; ERK, extracellular signal–regulated kinase; Gab, growth factor receptor–bound protein 2–associated binding protein; IMAC, immobilized metal affinity chromatography; IRS, insulin receptor substrate; MAPK, mitogen-activated protein kinase; MS/MS, tandem mass spectrometry; PI3K, phosphatidylinositol 3-kinase; PTRF, polymerase I and transcript release factor; pTyr, phosphorylated tyrosine; SHC, Src homology 2 domain containing transforming protein C; SHP-2, protein tyrosine phosphatase, non-receptor type II
ABSTRACT
The insulin-signaling network regulates blood glucose levels, controls metabolism, and when dysregulated, may lead to the development of type 2 diabetes. Although the role of tyrosine phosphorylation in this network is clear, only a limited number of insulin-induced tyrosine phosphorylation sites have been identified. To address this issue and establish temporal response, we have, for the first time, carried out an extensive, quantitative, mass spectrometry-based analysis of tyrosine phosphorylation in response to insulin. The study was performed with 3T3-L1 adipocytes stimulated with insulin for 0, 5, 15, and 45 min. It has resulted in the identification and relative temporal quantification of 122 tyrosine phosphorylation sites on 89 proteins. Insulin treatment caused a change of at least 1.3-fold in tyrosine phosphorylation on 89 of these sites. Among the responsive sites, 20 were previously known to be tyrosine phosphorylated with insulin treatment, including sites on the insulin receptor and insulin receptor substrate-1. The remaining 69 responsive sites have not previously been shown to be altered by insulin treatment. They were on proteins with a wide variety of functions, including components of the trafficking machinery for the insulin-responsive glucose transporter GLUT4. These results show that insulin-elicited tyrosine phosphorylation is extensive and implicate a number of hitherto unrecognized proteins in insulin action.
Metabolic control is primarily regulated by the insulin-signaling network. In healthy individuals, insulin stimulates glucose uptake from the bloodstream into adipose tissue and skeletal muscle while inhibiting glucose production in the liver. Dysregulation of this network associated with insulin resistance causes an increase in blood glucose and lipid levels, often initially associated with an increase in insulin levels and eventually culminating in type 2 diabetes (1). Understanding the signaling network activated by insulin stimulation is crucial for identifying the causes and effects of network dysregulation and insulin resistance.
Insulin binds to the insulin receptor at the cell surface and activates its tyrosine kinase activity, leading to autophosphorylation and phosphorylation of several receptor substrates. Phosphorylation of selected tyrosine sites on receptor substrates is known to activate different pathways leading to increased glucose uptake, lipogenesis, and glycogen and protein synthesis, as well as to stimulation of cell growth (1,2). In addition to activation of these pathways by tyrosine phosphorylation, several mechanisms of downregulating the response to insulin stimulation have also been identified. For instance, serine phosphorylation on insulin receptor substrate (IRS)-1 induced by a variety of factors has been shown to interfere with the activating effects of tyrosine phosphorylation by decreasing binding to the insulin receptor or increasing degradation of IRS-1 (1,3,4). Ser/Thr phosphorylation of the insulin receptor has also been shown to decrease tyrosine kinase activity (1). Downregulation of the insulin receptor (1) and IRS-1 (3) are two additional mechanisms of mediating insulin resistance. Many of these factors are reflected in decreased amounts of the tyrosine-phosphorylated receptor and receptor substrates with concomitant reduction in downstream signaling. Even though tyrosine phosphorylation plays a key role in insulin signaling, rather limited knowledge of specific phosphorylation sites, mainly focusing on tyrosine phosphorylation on the insulin receptor and IRS-1, is available so far. Therefore, we expected that a more comprehensive analysis of tyrosine phosphorylation upon insulin stimulation would lead to further insights into the biology of the signaling network.
We have recently developed a mass spectrometric methodology for the identification and quantification of tyrosine phosphorylation sites on many proteins (5). Here we have applied this methodology to the analysis of insulin signaling in 3T3-L1 adipocytes stimulated with insulin for 0, 5, 15, or 45 min. Using this approach, we were able to identify and quantify the temporal dynamics of many previously described sites on the insulin receptor and several insulin receptor substrates, as well as many additional sites, both previously characterized and novel, on other proteins associated with insulin signaling. These include sites related to the mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) pathways. Moreover, this analysis also produced temporal phosphorylation profiles for a number of novel phosphorylation sites on proteins which have, so far, not been directly associated with insulin signaling, such as proteins in the machinery of GLUT4 trafficking. The results of this study show that the insulin-elicited increase in tyrosine phosphorylation is more widespread than previously known and identify many new sites to be explored for their specific roles in insulin action.
RESEARCH DESIGN AND METHODS
Cell culture, insulin stimulation, and cell lysis.
3T3-L1 fibroblasts from the American Type Culture Collection were carried as fibroblasts and differentiated into adipocytes, as described previously (6). Confluent 10-cm plates of adipocytes at day 7 after differentiation (1 x 107 cells per 10-cm plate) were washed with and incubated in serum-free Dulbecco's modified Eagle's medium with 1 mg/ml BSA for 16 h. Cells were then stimulated with 150 nmol/l insulin in this medium for 5, 15, and 45 min; nontreated cells were used as 0-min time point. For every time point, three plates were prepared. After rinsing with PBS, cells were lysed in 1.5 ml 8 mol/l urea containing 1 mmol/l Na3VO4 (7); the cell lysate was frozen in liquid nitrogen and stored at –70°C until further use.
Sample processing and peptide immunoprecipitation.
Protein concentration was determined by bicinchoninic acid protein concentration assay (Pierce). Proteins were reduced, alkylated, and digested with modified trypsin (Promega; enzyme-to-substrate ratio 1:50) (5). The digest was acidified to pH 2 with HCl, centrifuged, and the supernatant was filtered (Millex-HV filter, 0.45-μm pore size; Millipore), desalted, and fractionated on a C18 Sep-Pak Plus Cartridge (Waters). Peptides eluted with 25% acetonitrile in 0.1% acetic acid were lyophilized, labeled with iTRAQ (1/4 plate per condition), and combined, and phosphotyrosine (pTyr) peptide immunoprecipitation was performed with anti-pTyr antibodies, as previously described (5), in 30 mmol/l TrisCl, 30 mmol/l NaCl, pH 7.4 containing 0.4% Nonidet P40 Substitute (Fluka). For peptide immunoprecipitation, a mixture of P-Tyr-100 (18 μg; Cell Signaling no. 9411) and PT-66 (12 μg; Sigma P3300), previously coupled to Protein G Plus-Agarose beads (20 μl; Calbiochem), was used, with the peptides derived from one 10-cm plate in 0.45 ml. Bound peptides were eluted from the antibody with 70 μl glycine 100 mmol/l, pH 2.1.
Chromatography and liquid chromatography tandem mass spectrometry.
Since peptide immunoprecipitation with pan-specific anti-phosphotyrosine antibodies is subject to nonspecific binding (specifically, those peptides containing aromatic amino acids [e.g., tyrosine, phenylalanine, and tryptophan] appear to be preferentially enriched), further enrichment for phosphopeptides was performed using immobilized metal affinity chromatography (IMAC), as previously described (5). This tandem affinity strategy virtually eliminated nonspecifically retained peptides, such that almost all peptides in the final analysis contained phosphorylated tyrosine. Phosphopeptides were eluted from the IMAC column to a capillary precolumn (100 μm i.d., packed with 10 μm ODS-A [Kanematsu]), which was then connected to a capillary analytical column (50 μm i.d., packed with 10 cm of 5 μm ODS-AQ [YMC-Waters]) with an integrated, laser-pulled (Model P-2000; Sutter Instrument) electrospray ionization emitter tip (2-μm diameter) (8). Peptides were eluted (flow rate 20 nl/min) from the liquid chromatography column to the quadrupole time-of-flight mass spectrometer (QSTAR XL; Applied Biosystems) with the following gradient: 0 min: 0% B; 10 min: 13% B; 105 min: 42% B; 115 min: 60% B; 122 min: 100% B (solvent A = 0.2 mol/l acetic acid and solvent B = 70% acetonitrile, 0.2 mol/l acetic acid). Data were acquired in information-dependent acquisition mode, in which a full scan mass spectrum (2.5 s) was followed by tandem mass spectrometry (MS/MS) of the four most abundant ions (4 s each) of charge state 2–5 using mass exclusion time of 25 s.
Phosphopeptide sequencing and quantitative analysis.
MS/MS spectra were extracted and searched against a rodent (mouse and rat) protein database (NCBI) using ProQuant (Applied Biosystems) as described previously (5). Mass tolerance was set to 2.2 atomic mass units for precursor ions and 0.15 atomic mass units for fragment ions. Phosphotyrosine-containing peptides were manually validated and quantified. Peak areas for each of the four signature peaks (m/z: 114, 115, 116, and 117) were obtained and corrected according to the manufacturer's instructions to account for isotopic overlap. Only spectra with signature peaks below 1,500 counts were considered for quantification. To compensate for small differences in the sample amounts at each time point, the results were normalized to those for nonphosphorylated peptides of 10 abundant proteins present in the samples (online appendix Table 1 [available at http://diabetes.diabetesjournals.org]). Finally, all data were normalized by the 5-min sample.
The complete analysis of tyrosine phosphorylation was performed as described above three separate times, starting with lysates from separate 10-cm plates.
RESULTS
Overview of insulin-elicited tyrosine phosphorylation.
To quantify temporal dynamics of tyrosine phosphorylation in the insulin-signaling network, we have immunoprecipitated stable isotope-coded, tyrosine-phosphorylated peptides from 3T3-L1 adipocytes stimulated with insulin for either 0, 5, 15, or 45 min (Fig. 1). IMAC–liquid chromatography MS/MS analysis of the immunoprecipitated samples generated quantitative, temporal phosphorylation profiles for 126 peptides from 89 proteins. More than 80% of the peptides were quantified from at least two of the three biological replicates with an average SD of 10% for the three analyses (online appendix, Table 2).
All the sites identified and quantified in this study are listed in Table 1 according to their changes in phosphorylation after 5-min insulin stimulation relative to control (no insulin stimulation). Table 2 of the online appendix provides a complete presentation of the data. In the description of the results given below, if the supporting data are not presented in a table or figure, they can be found in Table 1 of the text and/or Table 2 of the online appendix.
An overview of the results is as follows. Of the 122 sites in 89 proteins that were analyzed, the phosphorylation level of 86 sites in 68 proteins increased by a factor of 1.3 or more in response to insulin, while three sites on three proteins decreased by a factor of 1.3 or more. The remaining 33 sites showed <1.3-fold change in phosphorylation after 5-min insulin treatment. Among the 89 sites that were altered in their extent of phosphorylation >1.3-fold in response to insulin, 38 sites have not previously been identified as sites of tyrosine phosphorylation in any context. Moreover, among the 51 responsive sites previously identified in any context, there were 20 sites previously known to undergo a change in tyrosine phosphorylation in response to insulin or the closely related IGF-I. The remaining 31 sites of tyrosine phosphorylation have been identified in other contexts but were not previously known to be affected by insulin treatment. Thus, overall we have identified 69 sites with a response of at least 1.3-fold that are either entirely novel or novel in the context of insulin action.
Especially notable in the dataset are 12 peptides that increased in phosphorylation by >10-fold following 5 min of insulin stimulation (Table 2). These included all the peptides containing sites identified on the insulin receptor itself. In addition, several sites on proteins related to the MAPK pathway were found in this group: Y1171 on IRS-1 and Y660 on growth factor receptor–bound protein 2–associated binding protein (Gab) 1; two sites with very similar sequences (pYI/LDLDL) that bind protein tyrosine phosphatase, non-receptor type II (SHP-2) and thereby participate in activation of the MAPK pathway (9,10); Y53 on Sprouty4, a site known to have inhibitory effect on MAPK activation (11); and the doubly phosphorylated (T202 and Y204, T185 and Y187) active forms of extracellular signal–regulated kinase (ERK) 1 and 2 (12). Three other phosphorylation sites were also found to have >10-fold increase in phosphorylation: Y618 on adaptor protein with pleckstrin homology and src homology 2 domains (APS), an adaptor protein linking the insulin receptor to Cbl binding (13); Y521 on Munc18c, a novel site on this protein, which is involved in the fusion of GLUT4 vesicles with the plasma membrane (14); and Y1640 on Cdc42bpb, a novel site on this serine kinase, which may act as a downstream effector of Cdc42 in cytoskeleton reorganization (15). In the following sections, we describe in more detail various sets of tyrosine phosphorylation sites.
The head of the pathway: insulin receptor and substrates of the insulin receptor.
Following ligand binding, the insulin receptor autophosphorylates on selected tyrosine residues, increasing kinase activity and recruiting adaptor proteins and substrates. We detected the singly and doubly phosphorylated forms of the catalytic loop of the kinase domain (including Y1175, 1179, and 1180) with strong increase in phosphorylation up to 5 min and slight decrease to 45 min. In vitro, the triply phosphorylated form has been reported to lead to full activation of the phosphotransferase activity; however, in vivo, the doubly phosphorylated form was found to be the major form (16). The COOH-terminal tyrosines 1345 and 1351 were also identified to be phosphorylated with a time profile comparable to the sites in the catalytic loop. Those two sites have been reported to be involved in the regulation of the phosphotransferase activity of the insulin receptor (17).
IRS-1, IRS-2, Src homology two domain containing transforming protein C (SHC), Gab1, and APS are insulin receptor substrate/scaffolding proteins, the tyrosine phosphorylation of which connects the activation of the insulin receptor to specific pathways (1,2,10,18,19). Mouse knockouts of IRS-1 and IRS-2 have marked phenotypes that show these two IRSs play particularly prominent roles in insulin signaling (20). As expected, insulin caused a marked increase in the phosphorylation of one or more tyrosines on each of these substrate/scaffolding proteins. In the case of IRS-1, we were able to monitor the phosphorylation of 4 (Y460, Y935, Y983, Y1171) of the 10 tyrosine phosphorylation sites previously reported for mouse, rat, or human IRS-1 (PhosphoSite). Three of the sites were in the pYXXM motif, which is known to bind to the SH2 domains of the regulatory subunit of PI3K, and, as noted above, the fourth (Y1171) is a binding site for SHP-2 (10). They all showed maximal phosphorylation after 5-min insulin stimulation followed by a slight decrease with longer insulin stimulation. The peptide containing phosphorylation of Y460 demonstrated a greater decrease (to about 50% of maximum level), which was most likely due to the appearance of the doubly phosphorylated form (T448/Y460) with maximum intensity after 15-min stimulation. The T448 site has not previously been reported and its function is not known. However, as noted earlier, there are many phosphorylation sites on serine residues of IRS-1 that have been shown to alter its extent of tyrosine phosphorylation (3,4).
In the case of IRS-2, we identified and quantified six tyrosine phosphorylation sites: Y594, Y628, Y649, Y671, Y734, Y758, and Y814. Remarkably, although the sites of tyrosine phosphorylation on IRS-2 have been inferred by comparison to those on IRS-1 (21), to our knowledge, they have not been previously determined. All but one of these sites is in the motif pYXXM, which binds to SH2 domains of the regulatory subunit of PI3K. For all of them, with the exception of Y671, phosphorylation increased more than twofold from 0 to 5 min followed by mostly constant levels up to 45 min. Y671 showed only slight increase from 0 to 5 min, and Y734 decreased slightly from 5 to 45 min. However, the peptide containing Y734 was also observed in the doubly phosphorylated form with a serine phosphorylation at position 727 or 728; the doubly phosphorylated peptide slightly increased (20%) from 5 to 45 min, potentially offsetting the decrease of the singly phosphorylated form. As described above, two other substrate/scaffolding proteins, Gab1 and APS, showed a marked increase in tyrosine phosphorylation. In addition, there was an eightfold increase in the tyrosine phosphorylation of SHC, which generates a motif that binds to the adaptor protein Grb2 and contributes to activation of the MAPK pathway (22).
Proteins associated with the insulin receptor and/or the IRSs.
In addition to the substrates of the insulin receptor described above, insulin treatment caused increases in tyrosine phosphorylation on a number of other proteins known to be associated with the insulin receptor and/or its scaffolding substrates and to participate in insulin signaling. The tyrosine phosphatases, hemopoietic cell phosphatase, which can associate directly with the insulin receptor, and SHP-2, which associates with the IRSs and Gab1, underwent tyrosine phosphorylation on previously identified sites (23,24). The adaptor protein, Crk, and, to a small extent, the related adaptor protein CrkL, were tyrosine phosphorylated on previously undescribed sites. Crk and CrkL are known to associate with both the insulin receptor and the IRSs (25,26). The -type 85 kDa regulatory subunit of PI3K exhibited 1.5-fold increase in tyrosine phosphorylation of Y467 and Y580. Y580, but not Y467 (note that this peptide could be derived from either p85- or p55-), has previously been reported to be phosphorylated by the insulin receptor (27). PI3K binds through the SH2 domains on its regulatory subunit to the IRSs and is thereby activated to produce the key signaling lipid, PtdIns 3,4,5-trisphosphate, which in turn activates the kinase Akt (1,2). SHIP2, a phosphatase for the five phosphoryl group on the signaling lipid PtdIns 3,4,5 trisphosphate, showed an eightfold increase in phosphorylation on Tyr 887, another site not previously identified as responsive to insulin. SHIP2 binds to SHC, and may participate in shutting down insulin signaling through the PI3K/Akt pathway (28). Nck 1 and 2, closely related adaptor proteins that bind to IRS-1 and may serve as a link to the cytoskeleton (29,30), increased in tyrosine phosphorylation on a single site 1.8-fold. This site on Nck has previously been identified to undergo phosphorylation in response to epidermal growth factor but not insulin (5). Fer, a cytoplasmic tyrosine kinase that associates with IRS-1 in insulin-treated adipocytes (31), exhibited a 1.7-fold increase in phosphorylation on Y402. Finally, in contrast to the other proteins in this section, the Src family tyrosine kinase Fyn, which associates with tyrosine phosphorylated IRS-1 (32), underwent a 40% decrease in phosphorylation on Y417 over the 45 min of insulin exposure. This site is due to autophosphorylation and increases kinase activity (33), so dephosphorylation may be associated with reduced Fyn kinase activity. (The peptide containing this site is also found in three other Src tyrosine kinase family members, Yes, Src, and Lck. We have assigned this site to Fyn based on previous reports linking Fyn to IRS-1, but it is possible that the quantification of this site may reflect the sum total of these Src family members.)
Other known insulin-elicited phosphotyrosine proteins in adipocytes.
Among the pTyr proteins, there is a group that has previously been identified as undergoing insulin-stimulated tyrosine phosphorylation in adipocytes but whose roles in signaling, if any, are less clear. This group consists of caveolin 1 and 2 (34,35), polymerase I and transcript release factor (PTRF) (36), Syncrip (37), and fatty acid binding protein 4 (also known as aP2) (38). Caveolin 1 and 2 and PTRF are major protein components of the caveolar regions in the plasma membrane, which also contain insulin receptors (34,39). In this regard, we also found a novel insulin-elicited phosphorylation on another caveolar protein, known as sdr. Sdr has previously been found to be the protein in caveolae that binds protein kinase C (40). The sites of insulin-stimulated phosphorylation on caveolin 1/2 and fatty acid binding protein 4, but not those on PTRF and Syncrip, have been previously identified.
Proteins not previously associated with the insulin-signaling network.
In addition to the many proteins and phosphorylation sites which have been associated with insulin signaling, a substantial portion of the phosphorylation sites are on proteins that have not previously been reported to be directly involved in insulin signaling. In this section, we describe some, but not all, of these.
A major effect of insulin in adipocytes is the stimulation of glucose transport. The basis for this effect is the rapid introduction of additional glucose transporters of the GLUT4 type into the plasma membrane in response to insulin. The latter is achieved via the insulin-triggered movement of specialized intracellular vesicles containing GLUT4 to the plasma membrane and their fusion therewith (41). This process is referred to as GLUT4 translocation. Thus, it is of considerable interest that we found four proteins previously implicated in GLUT4 translocation, Syntaxin4, Munc18c, EH domain-containing protein 2, and Annexin II, to be tyrosine phosphorylated in response to insulin (Table 1 and Fig. 2). Syntaxin4 is the plasma membrane snare protein that associates with the snare VAMP2 on GLUT4 vesicles (41). Munc18c associates with syntaxin4 and thereby may inhibit the binding of GLUT4 vesicles to syntaxin4 in the absence of insulin (14). EH domain-containing protein 2 has been found to associate with GLUT4 (42), and it participates in GLUT4 endocytosis (43). Annexin II appears to be involved in GLUT4 translocation in a way that is not yet defined (44). Current evidence indicates that the signaling pathway to GLUT4 translocation proceeds from the insulin receptor through PI3K and the serine kinase Akt, which is not known to stimulate any tyrosine kinase (45). Thus, it is unclear whether any of these tyrosine phosphorylation sites are part of the signaling network leading to GLUT4 translocation, but because they are on proteins of the trafficking machinery, they deserve consideration.
Two transporters, a K-Cl cotransporter (Slc12a4) and an amino acid transporter (Slc38a2, SAT2), were tyrosine phosphorylated in response to insulin. Insulin has been reported to stimulate both of these transport systems. However, in the case of the K-Cl transporter, the basis for the stimulation appears to be an increase in transporter mRNA (46), and in the case of the amino acid transporter, translocation from an internal source, similar to that for GLUT4, has been found to underlie the stimulation (47). It may be that tyrosine phosphorylation also contributes to the stimulation.
A group of four of the stimulated pTyr proteins may participate in adhesions between adipocytes. These are three membrane-associated guanylate kinases (MPP1, MPP7, and SAP102), members of a family of proteins known to be involved in cell-to-cell adhesion (48), and the adipocyte adhesion protein, a transmembrane protein that has been shown to participate in cell aggregation (49). Thus, it is possible that insulin treatment may alter the adipocyte/adipocyte interactions.
Temporal dynamics of phosphorylation in response to insulin stimulation.
Sites that increase in phosphorylation throughout the time course (Fig. 4B) also cluster into proteins with similarities. There are two likely RNA-binding proteins, Syncrip and Hdlbp, and there are three membrane-associated guanylate kinases, Mpp 1 and 7 and SAP102. In addition, annexin II, which is in this group, is similar to the membrane-associated guanylate kinases in that all are peripheral membrane proteins associated with the plasma membrane (48,52).
It is interesting that sites with similar temporal profiles appear to group together in modules within the signaling network. We have made the same observation in the epidermal growth factor receptor signaling network (5). Such temporal analysis has the potential to reveal functions for previously uncharacterized sites and proteins.
DISCUSSION
This study is, to date, the most comprehensive analysis of the effect of insulin stimulation on site-specific protein tyrosine phosphorylation. We chose to use 3T3-L1 adipocytes for the analysis, since this cell type has been extensively studied, is highly insulin responsive, and is a model for the animal fat cell. Cells were continuously exposed to insulin for 0, 5, 15, or 45 min. Phosphotyrosine peptide immunoprecipitation combined with iTRAQ stable isotope labeling and IMAC–liquid chromatography MS/MS produced quantitative temporal phosphorylation profiles for a large proportion of the sites and proteins in the insulin-signaling network. For instance, we detected phosphorylation on many of the previously known sites of insulin-elicited tyrosine phosphorylation on known proteins. Identification of these sites provides validation of the method. It is worth noting that not every previously known site was detected in this analysis, due to a variety of reasons, including incompatibility with IMAC–liquid chromatography MS/MS analysis or low signal level for a given peptide. In several cases (e.g., c-Cbl Y369, Gab-1 Y407), sites were identified but could not be quantified due to low signal level for iTRAQ marker ions or interfering, coeluting peptides. In addition to previously known sites, we found 66 sites with at least 1.3-fold increase in phosphorylation upon 5-min insulin stimulation that had not previously been known to be insulin responsive. Thus, insulin-elicited tyrosine phosphorylation is much more extensive than previously known.
The significance for insulin action of tyrosine phosphorylation on each of the newly discovered sites remains to be determined. The present study offers many encouraging leads, such as the tyrosine phosphorylation of a group of proteins that are components in the trafficking machinery for GLUT4. The elucidation of the function of each tyrosine phosphorylation will require an in-depth investigation of the particular site on the particular protein. A general approach to this problem is to knock down the protein in question, replace it with mutant protein containing nonphosphorylatable phenylalanine in place of tyrosine, and examine the effect of the replacement on insulin action. Clearly, further characterization of these novel sites and proteins will result in a significant expansion of our knowledge of the insulin-signaling network.
In the future, application of this approach to the investigation of insulin signaling in various tissues of normal and diabetic mice, such as adipose tissue, muscle, and liver, will be a powerful approach to identify tissue-specific sites of tyrosine phosphorylation and the specific alterations in tyrosine phosphorylation in the diabetic state.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health Grant DK42816.
We thank other members of the White lab and Sampsa Hautaniemi in the Lauffenburger lab at MIT for their assistance and helpful discussions.
FOOTNOTES
Additional information for this article can be found in an online appendix at http://diabetes.diabetesjournals.org.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Pirola L, Johnston AM, Van Obberghen E: Modulation of insulin action. Diabetologia 47:170–184, 2004
Saltiel AR, Kahn CR: Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414:799–806, 2001
Gual P, Le Marchand-Brustel Y, Tanti JF: Positive and negative regulation of insulin signaling through IRS-1 phosphorylation. Biochimie 87:99–109, 2005
Zick Y: Uncoupling insulin signalling by serine/threonine phosphorylation: a molecular basis for insulin resistance. Biochem Soc Trans 32:812–816, 2004
Zhang Y, Wolf-Yadlin A, Ross PL, Pappin DJ, Rush J, Lauffenburger DA, White FM: Time-resolved mass spectrometry of tyrosine phosphorylation sites in the epidermal growth factor receptor signaling network reveals dynamic modules. Mol Cell Proteomics 4:1240–1250, 2005
Frost SC, Lane MD: Evidence for the involvement of vicinal sulfhydryl groups in insulin-activated hexose transport by 3T3–L1 adipocytes. J Biol Chem 260:2646–2652, 1985
Gordon JA: Use of vanadate as protein-phosphotyrosine phosphatase inhibitor. Methods Enzymol 201:477–482, 1991
Martin SE, Shabanowitz J, Hunt DF, Marto JA: Subfemtomole MS and MS/MS peptide sequence analysis using nano-HPLC micro-ESI fourier transform ion cyclotron resonance mass spectrometry. Anal Chem 72:4266–4274, 2000
Cunnick JM, Mei L, Doupnik CA, Wu J: Phosphotyrosines 627 and 659 of Gab1 constitute a bisphosphoryl tyrosine-based activation motif (BTAM) conferring binding and activation of SHP2. J Biol Chem 276:24380–24387, 2001
White MF: IRS proteins and the common path to diabetes. Am J Physiol Endocrinol Metab 283:E413–E422, 2002
Kim HJ, Bar-Sagi D: Modulation of signalling by Sprouty: a developing story. Nat Rev Mol Cell Biol 5:441–450, 2004
Her JH, Lakhani S, Zu K, Vila J, Dent P, Sturgill TW, Weber MJ: Dual phosphorylation and autophosphorylation in mitogen-activated protein (MAP) kinase activation. Biochem J 296:25–31, 1993
Ahn MY, Katsanakis KD, Bheda F, Pillay TS: Primary and essential role of the adaptor protein APS for recruitment of both c-Cbl and its associated protein CAP in insulin signaling. J Biol Chem 279:21526–21532, 2004
Kanda H, Tamori Y, Shinoda H, Yoshikawa M, Sakaue M, Udagawa J, Otani H, Tashiro F, Miyazaki J, Kasuga M: Adipocytes from Munc18c-null mice show increased sensitivity to insulin-stimulated GLUT4 externalization. J Clin Invest 115:291–301, 2005
Leung T, Chen XQ, Tan I, Manser E, Lim L: Myotonic dystrophy kinase-related Cdc42-binding kinase acts as a Cdc42 effector in promoting cytoskeletal reorganization. Mol Cell Biol 18:130–140, 1998
White MF, Shoelson SE, Keutmann H, Kahn CR: A cascade of tyrosine autophosphorylation in the beta-subunit activates the phosphotransferase of the insulin receptor. J Biol Chem 263:2969–2980, 1988
Tennagels N, Bergschneider E, Al-Hasani H, Klein HW: Autophosphorylation of the two C-terminal tyrosine residues Tyr1316 and Tyr1322 modulates the activity of the insulin receptor kinase in vitro. FEBS Lett 479:67–71, 2000
Holgado-Madruga M, Emlet DR, Moscatello DK, Godwin AK, Wong AJ: A Grb2-associated docking protein in EGF- and insulin-receptor signalling. Nature 379:560–564, 1996
Ahmed Z, Pillay TS: Functional effects of APS and SH2-B on insulin receptor signalling. Biochem Soc Trans 29:529–534, 2001
Biddinger SB, Kahn CR: From mice to men: insights into the insulin resistance syndromes. Annu Rev Physiol 68:123–158 2005
Sun XJ, Wang LM, Zhang Y, Yenush L, Myers MG Jr, Glasheen E, Lane WS, Pierce JH, White MF: Role of IRS-2 in insulin and cytokine signalling. Nature 377:173–177, 1995
Sasaoka T, Kobayashi M: The functional significance of Shc in insulin signaling as a substrate of the insulin receptor. Endocr J 47:373–381, 2000
Uchida T, Matozaki T, Noguchi T, Yamao T, Horita K, Suzuki T, Fujioka Y, Sakamoto C, Kasuga M: Insulin stimulates the phosphorylation of Tyr538 and the catalytic activity of PTP1C, a protein tyrosine phosphatase with Src homology-2 domains. J Biol Chem 269:12220–12228, 1994
Stein-Gerlach M, Kharitonenkov A, Vogel W, Ali S, Ullrich A: Protein-tyrosine phosphatase 1D modulates its own state of tyrosine phosphorylation. J Biol Chem 270:24635–24637, 1995
Feller SM: Crk family adaptors-signalling complex formation and biological roles. Oncogene 20:6348–6371, 2001
Klammt J, Barnikol-Oettler A, Kiess W: Mutational analysis of the interaction between insulin receptor and IGF-I receptor with c-Crk and Crk-L in a yeast two-hybrid system. Biochem Biophys Res Commun 325:183–190, 2004
Hayashi H, Nishioka Y, Kamohara S, Kanai F, Ishii K, Fukui Y, Shibasaki F, Takenawa T, Kido H, Katsunuma N, et al.: The alpha-type 85-kDa subunit of phosphatidylinositol 3-kinase is phosphorylated at tyrosines 368, 580, and 607 by the insulin receptor. J Biol Chem 268:7107–7117, 1993
Dyson JM, Kong AM, Wiradjaja F, Astle MV, Gurung R, Mitchell CA: The SH2 domain containing inositol polyphosphate 5-phosphatase-2: SHIP2. Int J Biochem Cell Biol 37:2260–2265, 2005
Tu Y, Liang L, Frank SJ, Wu C: Src homology 3 domain-dependent interaction of Nck-2 with insulin receptor substrate-1. Biochem J 354:315–322, 2001
Rivera GM, Briceno CA, Takeshima F, Snapper SB, Mayer BJ: Inducible clustering of membrane-targeted SH3 domains of the adaptor protein Nck triggers localized actin polymerization. Curr Biol 14:11–22, 2004
Iwanishi M, Czech MP, Cherniack AD: The protein-tyrosine kinase fer associates with signaling complexes containing insulin receptor substrate-1 and phosphatidylinositol 3-kinase. J Biol Chem 275:38995–39000, 2000
Sun XJ, Pons S, Asano T, Myers MG Jr, Glasheen E, White MF: The Fyn tyrosine kinase binds Irs-1 and forms a distinct signaling complex during insulin stimulation. J Biol Chem 271:10583–10587, 1996
Roskoski R Jr: Src protein-tyrosine kinase structure and regulation. Biochem Biophys Res Commun 324:1155–1164, 2004
Kimura A, Mora S, Shigematsu S, Pessin JE, Saltiel AR: The insulin receptor catalyzes the tyrosine phosphorylation of caveolin-1. J Biol Chem 277:30153–30158, 2002
Lee H, Park DS, Wang XB, Scherer PE, Schwartz PE, Lisanti MP: Src-induced phosphorylation of caveolin-2 on tyrosine 19: phospho-caveolin-2 (Tyr(P)19) is localized near focal adhesions, remains associated with lipid rafts/caveolae, but no longer forms a high molecular mass hetero-oligomer with caveolin-1. J Biol Chem 277:34556–34567, 2002
Ibarrola N, Molina H, Iwahori A, Pandey A: A novel proteomic approach for specific identification of tyrosine kinase substrates using [13C]tyrosine. J Biol Chem 279:15805–15813, 2004
Hresko RC, Mueckler M: Identification of pp68 as the tyrosine-phosphorylated form of SYNCRIP/NSAP1: a cytoplasmic RNA-binding protein. J Biol Chem 277:25233–25238, 2002
Hresko RC, Hoffman RD, Flores-Riveros JR, Lane MD: Insulin receptor tyrosine kinase-catalyzed phosphorylation of 422(aP2) protein: substrate activation by long-chain fatty acid. J Biol Chem 265:21075–21085, 1990
Aboulaich N, Vainonen JP, Stralfors P, Vener AV: Vectorial proteomics reveal targeting, phosphorylation and specific fragmentation of polymerase I and transcript release factor (PTRF) at the surface of caveolae in human adipocytes. Biochem J 383:237–248, 2004
Mineo C, Ying YS, Chapline C, Jaken S, Anderson RG: Targeting of protein kinase Calpha to caveolae. J Cell Biol 141:601–610, 1998
Watson RT, Kanzaki M, Pessin JE: Regulated membrane trafficking of the insulin-responsive glucose transporter 4 in adipocytes. Endocr Rev 25:177–204, 2004
Park SY, Ha BG, Choi GH, Ryu J, Kim B, Jung CY, Lee W: EHD2 interactswith the insulin-responsive glucose transporter (GLUT4) in rat adipocytes and may participate in insulin-induced GLUT4 recruitment. Biochemistry 43:7552–7562, 2004
Guilherme A, Soriano NA, Furcinitti PS, Czech MP: Role of EHD1 and EHBP1 in perinuclear sorting and insulin-regulated GLUT4 recycling in 3T3–L1 adipocytes. J Biol Chem 279:40062–40075, 2004
Huang J, Hsia SH, Imamura T, Usui I, Olefsky JM: Annexin II is a thiazolidinedione-responsive gene involved in insulin-induced glucose transporter isoform 4 translocation in 3T3–L1 adipocytes. Endocrinology 145:1579–1586, 2004
Welsh GI, Hers I, Berwick DC, Dell G, Wherlock M, Birkin R, Leney S, Tavare JM: Role of protein kinase B in insulin-regulated glucose uptake. Biochem Soc Trans 33:346–349, 2005
Shen MR, Lin AC, Hsu YM, Chang TJ, Tang MJ, Alper SL, Ellory JC, Chou CY: Insulin-like growth factor 1 stimulates KCl cotransport, which is necessary for invasion and proliferation of cervical cancer and ovarian cancer cells. J Biol Chem 279:40017–40025, 2004
Hyde R, Peyrollier K, Hundal HS: Insulin promotes the cell surface recruitment of the SAT2/ATA2 system A amino acid transporter from an endosomal compartment in skeletal muscle cells. J Biol Chem 277:13628–13634, 2002
Funke L, Dakoji S, Bredt DS: Membrane-associated guanylate kinases regulate adhesion and plasticity at cell junctions. Annu Rev Biochem 74:219–245, 2005
Eguchi J, Wada J, Hida K, Zhang H, Matsuoka T, Baba M, Hashimoto I, Shikata K, Ogawa N, Makino H: Identification of adipocyte adhesion molecule (ACAM), a novel CTX gene family, implicated in adipocyte maturation and development of obesity. Biochem J 387:343–353, 2005
Kayali AG, Austin DA, Webster NJ: Stimulation of MAPK cascades by insulin and osmotic shock: lack of an involvement of p38 mitogen-activated protein kinase in glucose transport in 3T3–L1 adipocytes. Diabetes 49:1783–1793, 2000
Waters SB, Yamauchi K, Pessin JE: Insulin-stimulated disassociation of the SOS-Grb2 complex. Mol Cell Biol 15:2791–2799, 1995
Gerke V, Moss SE: Annexins and membrane dynamics. Biochim Biophys Acta 1357:129–154, 1997(Katrin Schmelzle, Susan K)