Biochemical Journal

Research article

Dock/Nck facilitates PTP61F/PTP1B regulation of insulin signalling

Chia-Lun Wu, Bree Buszard, Chun-Hung Teng, Wei-Lin Chen, Coral G. Warr, Tony Tiganis, Tzu-Ching Meng

Abstract

PTP1B (protein tyrosine phosphatase 1B) is a negative regulator of IR (insulin receptor) activation and glucose homoeostasis, but the precise molecular mechanisms governing PTP1B substrate selectivity and the regulation of insulin signalling remain unclear. In the present study we have taken advantage of Drosophila as a model organism to establish the role of the SH3 (Src homology 3)/SH2 adaptor protein Dock (Dreadlocks) and its mammalian counterpart Nck in IR regulation by PTPs. We demonstrate that the PTP1B orthologue PTP61F dephosphorylates the Drosophila IR in S2 cells in vitro and attenuates IR-induced eye overgrowth in vivo. Our studies indicate that Dock forms a stable complex with PTP61F and that Dock/PTP61F associate with the IR in response to insulin. We report that Dock is required for effective IR dephosphorylation and inactivation by PTP61F in vitro and in vivo. Furthermore, we demonstrate that Nck interacts with PTP1B and that the Nck/PTP1B complex inducibly associates with the IR for the attenuation of IR activation in mammalian cells. Our studies reveal for the first time that the adaptor protein Dock/Nck attenuates insulin signalling by recruiting PTP61F/PTP1B to its substrate, the IR.

  • Dock/Nck
  • Drosophila
  • insulin receptor
  • insulin signalling
  • protein tyrosine phosphatase 61F/protein tyrosine phosphatase 1B (PTP61F/PTP1B)
  • tyrosine phosphorylation

INTRODUCTION

Insulin is a key anabolic hormone that promotes the storage of fuels as macromolecules. Insulin exerts its effects by binding to its receptor, a PTK (protein tyrosine kinase) that undergoes autophosphorylation and activation and phosphorylates key substrates, including IRS [IR (insulin receptor) substrate] proteins 1–4 [1,2]. Tyrosine phosphorylation of IRS scaffold proteins allows for the nucleation of signalling complexes and the activation of key pathways, including the PI3K (phosphatidylinositol 3-kinase)/Akt pathway that elicits many of insulin's metabolic and mitogenic effects [1,2]. In mammals, a key function of insulin is to lower blood glucose levels after a meal. Insulin serves to promote glucose uptake in muscle and fat and suppress hepatic glucose production [1,2]. Drosophila melanogaster is an ideal model organism for investigating the signalling networks induced by insulin [36]. Indeed, many of the key pathway components and modulators of IR signalling were first described in flies. IR signalling regulates varied biological responses in flies, including longevity, female fertility and visual system formation [37].

PTPs (protein tyrosine phosphatases) are a large and structurally diverse family of enzymes that are integral to the spatial and temporal control of receptor PTK phosphorylation, activation and signalling [8,9]. The prototypic family member PTP1B is a negative regulator of IR activation and signalling [9]. In particular, PTP1B dephosphorylates the IR tandem Tyr1162/Tyr1163 PTK activation loop tyrosine phosphorylation site that is necessary for IR activation and signalling [1012]. In keeping with PTP1B being an integral negative regulator of IR signalling, previous studies have shown that PTP1B-null (Ptpn1−/−) mice exhibit enhanced insulin sensitivity associated with increased IR activation in liver and muscle [1216]. PTP1B deficiency improves glucose tolerance and delays the onset of diabetes in IRS-2-null mice [17] and in mice that are heterozygous deficient for both IR and IRS-1 [18], whereas transgenic overexpression of PTP1B in muscle promotes insulin resistance [19]. Moreover, PTP1B expression and/or activity may be increased in muscle and adipose tissue in insulin-resistant humans and rodents [2023]. PTP1B's integral role in insulin signalling makes it an exciting therapeutic target for the alleviation of insulin resistance in Type 2 diabetes [23,24].

Although PTPs such as PTP1B display exquisite substrate selectivity in vivo, it remains incompletely understood how this selectivity is conferred. Although the catalytic domain of PTP1B has inherent specificity for tyrosyl phosphorylated substrates, additional factors contribute to substrate selectivity [9,25,26]. PTP1B is targeted to the cytoplasmic face of the ER (endoplasmic reticulum) by a hydrophobic C-terminus [27], and this subcellular compartmentalization has spatial and temporal implications for PTP1B substrate selectivity. In particular, PTP1B's location in the ER may limit access to substrates such as the EGFR (epidermal growth factor receptor) at the plasma membrane. Indeed, PTP1B is thought to efficiently dephosphorylate the tyrosine-phosphorylated EGFR only after the receptor PTK is endocytosed [28]. The non-catalytic C-terminus of PTP1B also contains a proline-rich region that allows PTP1B to interact with SH3 (Src homology 3) domains to dephosphorylate substrates such as p130cas and c-Src [2933]. Two proline residues in the C-terminus of PTP1B interact with the SH3 domain of p130cas for efficient p130cas and c-Src dephosphorylation to regulate integrin signalling in fibroblasts in vitro [29,32,33]. Thus the proline-rich region in PTP1B may be important in conferring substrate selectivity.

The putative PTP1B orthologue in Drosophila melanogaster is PTP61F [34]. Similar to its mammalian counterpart, PTP61F is ubiquitous and expressed at all stages of development, including oogenesis and embryogenesis [35,36]. Alternative splicing of PTP61F message can result in two variants: one that is targeted similarly to PTP1B [27] to intracellular membranes, including the ER, by a hydrophobic C-terminal tail [34] and another that lacks the hydrophic C-terminus and is targeted to the nucleus by a nuclear localization sequence [34], reminiscent of mammalian TCPTP (T-cell PTP) that shares 74% catalytic domain sequence identity with PTP1B [9,37]. Hereafter, only the ER-targeted form of PTP61F will be referred to. Similar to PTP1B, PTP61F has a proline-rich sequence (five PxxP motifs) in its non-catalytic C-terminus [34]. PTP61F has been shown to interact with the adaptor protein Dock (Dreadlocks), which is composed of three SH3 domains and a single SH2 domain [38,39]. Indeed, Dock was first identified as a protein that interacted with PTP61F in a yeast two-hybrid screen [40]. This interaction was confirmed in co-immunoprecipitation experiments from Drosophila embryo extracts and Schneider II (S2) cells and shown to be mediated by one or more of the Dock SH3 domains and the PTP61F non-catalytic C-terminus [40]. Interestingly, there is also biochemical and genetic evidence that Dock interacts with the IR. Dock is required for photoreceptor-cell axon guidance and is localized to photoreceptor-cell growth cones [38]. The Drosophila IR is also enriched in photoreceptor-cell growth cones and is similarly required for photoreceptor-cell axon guidance [41]. Dock has been shown to interact genetically with the IR in photoreceptor cell axon guidance and this interaction is dependent on IR PTK activity [41]. Dock has been shown to bind to the IR C-terminal tail [41] that contains tyrosine phosphorylation sites and proline-rich sequences [41].

In the present study we assess the role of the adaptor Dock in IR regulation by PTP61F. We report the Drosophila IR is a substrate for PTP61F and that Dock is required for efficient IR dephosphorylation in vitro and in vivo. Moreover, we report that the mammalian Dock counterpart, Nck, may similarly be required for IR dephosphorylation by PTP1B. Thus our studies provide evidence that adaptor proteins play an important role in conferring PTP substrate selectivity.

MATERIALS AND METHODS

Reagents

Rabbit anti-Dock and mouse anti-HA (haemagglutinin; 12CA5) and anti-c-Myc (9E10) antibodies were from Abcam; recombinant human insulin, mouse anti-tubulin and anti-FLAG M2 antibodies were from Sigma–Aldrich; rabbit anti-phospho-Tyr1162/Tyr1163-IRβ antibody was from Biosource International; rabbit anti-IRβ and mouse anti-actin antibodies were from Santa Cruz Biotechnology; 21-nucleotide siRNA (small interfering RNA) duplexes to human NCK1 were from Thermo Scientific; and mouse anti-NCK1 antibody was from BD Biosciences. The mouse monoclonal anti-PTP1B antibody (FG6) was provided by Professor N.K. Tonks (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, U.S.A.) and the PTP61F antibody was provided by Professor J.E. Dixon (University of California, San Diego, La Jolla, CA, U.S.A.).

Plasmid constructs, recombinant PTP61F and substrate trapping

cDNA sequences corresponding to the full length or the C-terminally truncated [PTP61F1–339] PTP61F in the wild-type, the C237S phosphatase-dead mutant or the D203A trapping mutant form were constructed with an HA tag at the N-terminus and cloned into a pET-28a (for protein expression in Escherichia coli) or a pAc5.1A vector (for protein expression in S2 cells). cDNA sequence corresponding to full-length human NCK1 was purchased from Open Biosystems. For expression in mammalian cells, NCK1 was cloned into a pRK5 vector. Protein purification from E. coli extracts was described previously [42]. The wild-type form of PTP61F1–339 was characterized biochemically as an active phosphatase (results not shown). The in vitro substrate-trapping procedure has been described previously [42]. In brief, the HA-tagged wild-type or D203A mutant form of C-terminally truncated PTP61F recombinant protein was incubated with an aliquot of total lysate prepared from S2 cells that were stimulated with 200 nM insulin for 20 min. After incubation at 4 °C for 30 min, an aliquot of immobilized anti-HA antibody–agarose beads was added for an additional 3 h of incubation. After extensive washes, beads were boiled in SDS sample buffer, and the eluted proteins were analysed by immunoblotting with antibodies to the tandem Tyr1162/Tyr1163 phosphorylated and activated human IR β subunit [Phos-IR(Tyr1162/Tyr1163)], which also recognizes the phosphorylated and activated Drosophila IR [Phos-IR(Tyr1162/Tyr1163)].

Cell culture

Drosophila S2 cells were maintained in 1× Schneider medium supplemented with 10% FBS (fetal bovine serum) at 24 °C. For transient transfection with plasmid, Lipofectamine™ 2000 (Invitrogen) was used as a vehicle following the manufacturer's directions. For RNAi (RNA interference)-mediated knockdown in S2 cells, dsRNAs (double-stranded RNAs) of approximately 700 bp corresponding to the PTP61F and Dock genes were added to S2 cells incubated in serum-free culture medium as described previously [43,44]. After a 2 h treatment with dsRNA, cells were incubated in complete medium for 24–48 h and then processed for immunoprecipitation or immunoblotting as indicated. HEK-293 (human embryonic kidney 293) cells were routinely maintained in DMEM (Dulbecco's modified Eagle's medium) supplemented with 10% FBS. For overexpression studies, HEK-293 cells (5×105 cells/6 cm plate) were incubated with plasmid DNA (1 μg/6 cm plate) and Lipofectamine™ 2000, according to the manufacturer's directions. For RNAi studies, HEK-293 cells (5×105 cells/6 cm plate) were incubated with 100 nM GFP (green fluorescence protein; 5′-GCAGCUGACCCUGAAGUUCAU-3′) or Nck1 (5′-GAUUAUGGCUUCUGGAUGAUU-3′) siRNAs and Lipofectamine™ 2000, according to the manufacturer's directions. At 2 days after transfection, cells were deprived of serum for 6 h and then stimulated with insulin for indicated times.

Drosophila stocks and transgenics

GMR-gal4 and UAS (upstream activating sequence)-dInR strains were provided by H.E. Richardson (Peter MacCallum Cancer Center, Melbourne, Australia). UAS-Dock flies were provided by Larry Zipursky (Howard Hughes Medical Institute, University of California, Los Angeles, CA, U.S.A.) [39]. Dpp-gal4 flies were from the Bloomington Stock Centre. To generate UAS constructs for PTP61F, the coding region for PTP61F (the ER-targeted variant [34]) was amplified by RT (reverse transcription)–PCR from whole fly Canton S RNA. The fidelity of the coding region was verified by sequencing. The cDNA was cloned into the pNmyc-UAST vector (C.G. Warr, unpublished work) in-frame with the initiation codon and three copies of the Myc tag coding sequence. The resulting protein had the Myc tag at the N-terminus. Transgenic flies were generated by injection into w1118 embryos using standard procedures for P element-mediated transformation. Several independent transformant lines were established and analysed for each construct. Fly crosses were performed at 25 °C.

Drosophila eye microscopy

Flies were collected 3–5 days post-eclosion. Scanning electron microscopy was performed on a Hitachi S570 scanning electron microscope at 15 kV and ×100 magnification (Monash MicroImaging, Monash University). Images were captured using Spectrum NT software.

RESULTS

The IR is a substrate of PTP61F

The sequence and structural similarity between PTP61F and PTP1B [45] suggests that these phosphatases may recognize their substrates in an evolutionarily conserved manner. In mammalian cells, the IR is a bona fide substrate of PTP1B [9]. The PTP1B catalytic domain has a second phosphotyrosine-binding pocket that allows for the recognition of tandem phosphorylated substrates such as the Tyr1162/Tyr1163 phosphorylated IR [10,11,46]. Autophosphorylation of the IR activation loop Tyr1162/Tyr1163 site is required for IR activation in response to insulin [47]. The IR Tyr1162/Tyr1163 phosphorylation site is conserved in the Drosophila IR (Tyr1553/Tyr1554) and can be readily detected in Drosophila S2 cells in response to insulin with antibodies to the phosphorylation site in the human IR (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/439/bj4390151add.htm). To determine if PTP61F might have the capacity to regulate the IR in Drosophila, we first determined whether the catalytic domain of PTP61F [PTP61F1–339] could dephosphorylate the Drosophila IR in vitro. Purified recombinant HA-tagged PTP61F1–339 was incubated with lysates from insulin-stimulated S2 cells and the status of IR activation was assessed with antibodies to the human IR Tyr1162/Tyr1163 phosphorylation site. HA–PTP61F1–339 readily dephosphorylated the Tyr1553/Tyr1554 phosphorylated IR in S2 cell lysates (Figure 1A). To determine whether PTP61F could act directly on the tyrosine phosphorylated IR, we took advantage of the corresponding PTP61F D203A ‘substrate-trapping’ mutant; PTP ‘substrate-trapping’ mutants have the capacity to form direct and stable complexes with tyrosine phosphorylated substrates. Purified recombinant wild-type HA–PTP61F1–339 and HA–PTP61F1–339D203A proteins were incubated with insulin-stimulated S2 cell lysates and immunoprecipitated with HA antibodies, and the associated proteins were eluted and subjected to immunoblot analysis with antibodies to the phosphorylated IR (Figure 1B). The Tyr1553/Tyr1554 tyrosine phosphorylated IR was eluted from HA–PTP61F1–339D203A, but not HA–PTP61F1–339, immunoprecipitates (Figure 1B). Therefore, these results establish PTP61F's capacity to act on Drosophila IR directly. To determine whether PTP61F regulates the IR in a cellular context, we knocked down PTP61F expression by RNAi using dsRNAs [43,44]. Insulin-induced IR Tyr1553/Tyr1554 tyrosine phosphorylation in S2 cells was significantly increased after PTP61F knockdown (Figure 1C; see also Supplementary Figure S2 at http://www.BiochemJ.org/bj/439/bj4390151add.htm). Taken together, these results indicate that the IR can serve as a bona fide substrate for PTP61F.

Figure 1 PTP61F recognizes IR as a substrate

(A) S2 cells were treated with 200 nM recombinant human insulin for 10 min. Aliquots (25 μg) of total lysates were incubated with the recombinant purified HA–PTP61F1–339 and then resolved by SDS/PAGE and immunoblotted with antibodies to the phosphorylated and activated IR [Phos-IR (Y1549Y1550)] and tubulin. Rx, reaction. (B) Aliquots of total cell lysates prepared from insulin-treated S2 cells were incubated with either purified HA–PTP61F1–339 or HA–PTP61F1–339D203A. HA immunoprecipitates (IPs) were subjected to SDS/PAGE and processed for immunoblotting (top) or silver staining (bottom). (C) Endogenous PTP61F in S2 cells was knocked down with PTP61F dsRNA [43,44]. The dsRNA-treated S2 cells were then stimulated with 10 nM insulin for the indicated times. Aliquots of total lysates were resolved by SDS/PAGE and subjected to immunoblotting as indicated. Similar results for (AC) were observed in two independent experiments.

Overexpression of PTP61F suppresses IR-induced overgrowth in vivo

Next, we determined whether the interaction between PTP61F and the IR was of biological relevance. When the IR is overexpressed in the developing Drosophila eye using the GMR-GAL4 driver, it causes severe eye overgrowth due to hyperproliferation and increased cell size [48]. Previous studies have shown that co-expression of negative regulators of the insulin pathway can suppress this overgrowth phenotype [49]. To determine if PTP61F negatively regulates insulin signalling in vivo, we generated transgenic flies which carry the coding sequence of PTP61F under the control of the UAS regulatory sequence (UAS-PTP61F). Expression of PTP61F alone using the GMR-GAL4 driver had no effect on eye development (Figure 2). As reported previously [48], overexpression of the IR using the GMR-GAL4 driver led to a severe eye overgrowth phenotype, with substantial eye outgrowth and disorganization (Figure 2). Importantly, co-overexpression of PTP61F with the IR significantly reduced the IR-induced eye overgrowth (Figure 2). Taken together, our results indicate that PTP61F is an integral negative regulator of IR activation and signalling in flies.

Figure 2 PTP61F suppresses IR-driven eye overgrowth

Control flies (UAS-PTP61F/CyO; UAS-InR/+) and those expressing IR (GMR-Gal4/CyO; UAS-InR/+) or IR plus PTP61F (GMR-Gal4/UAS-PTP61F; UAS-InR/+) were processed for scanning electron microscopy (×150) as described in the Materials and methods section. Results shown are representative of two independent experiments (n=4–5 per genotype).

Dock bridges PTP61F and the IR and promotes IR dephosphorylation

Previous studies have shown that PTP61F interacts with Dock [3840] and that Dock interacts with the IR in flies [41]. Having established that the IR can serve as a PTP61F substrate, we next determined whether Dock might influence PTP61F's capacity to dephosphorylate the IR. First, we performed reciprocal co-immunoprecipitation experiments to confirm Dock association with PTP61F and to assess its dependence on insulin. We found that endogenous Dock in S2 cells readily associated with ectopically expressed and immunoprecipitated HA–PTP61F and this was independent of insulin stimulation (Figure 3A). Similarly, endogenous PTP61F co-precipitated with overexpressed Myc–Dock, with and without insulin stimulation (Figure 3B). Therefore Dock and PTP61F can form a stable complex and this can occur independently of insulin stimulation. We then asked whether Dock could recruit the tyrosine phosphorylated IR to the Dock–PTP61F complex for dephosphorylation by PTP61F. Myc–Dock alone or Myc–Dock and wild-type PTP61F or the phosphatase-dead ‘substrate-trapping’ mutant PTP61F-C237S were expressed in S2 cells. After insulin stimulation, Myc–Dock was immunoprecipitated and the level of phosphorylated IR associated with Dock, or Dock–PTP61F complexes, was examined. Tyr1553/Tyr1554 phosphorylated IR was readily detected in Myc–Dock alone precipitates (Figure 3C). HA–PTP61F also co-precipitated with Myc–Dock, but phosphorylated IR was not evident in precipitated Myc–Dock/HA–PTP61F complexes (Figure 3C). On the other hand, reproducibly greater amounts of Tyr1553/Tyr1554 phosphorylated IR were detected in Myc–Dock/HA–PTP61F-C237S complexes (Figure 3C). These results are consistent with Dock bridging PTP61F and its substrate the IR. Accordingly, we determined whether Dock could facilitate PTP-mediated IR dephosphorylation. We used RNAi to ablate Dock in S2 cells and assessed insulin-induced IR Tyr1553/Tyr1554 phosphorylation. We found that Dock knockdown significantly enhanced insulin-induced IR phosphorylation (Figure 3D). Taken together, these results indicate that Dock might serve to recruit PTP61F for the attenuation of IR phosphorylation and signalling.

Figure 3 Dock acts co-ordinately with PTP61F to regulate the IR

(A) S2 cells expressing HA-tagged full-length PTP61F were stimulated with insulin for 20 min. HA immunoprecipitates (IPs) were resolved by SDS/PAGE and processed for immunoblotting or silver staining as indicated. (B) S2 cells expressing Myc-tagged Dock were stimulated with 200 nM insulin for 20 min. Myc immunoprecipitates were resolved by SDS/PAGE and processed for immunoblotting or silver staining as indicated. (C) Myc–Dock was co-expressed with either the HA–PTP61F (WT) or HA–PTP61F-C237S mutant in S2 cells. Cells were stimulated with 200 nM insulin for 20 min and Myc immunoprecipitates were resolved by SDS-PAGE and processed for immunoblotting, or silver staining as indicated. Similar results for (AC) were observed in two independent experiments. In silver stain experiments, the IgG heavy chain (IgGHC) can be observed. (D) Endogenous Dock in S2 cells was knocked down with Dock dsRNA. As a control, S2 cells were treated with dsRNA for the gene encoding GFP. The dsRNA-treated S2 cells were stimulated after 48 h with 200 nM insulin for the indicated times. Aliquots of total lysates were resolved by SDS/PAGE and subjected to immunoblotting as indicated. The right-hand panel shows a densitometric analysis of phosphorylated IR normalized to tubulin. Units shown are arbitrary (A.U.) and are means±S.E.M. (n=3); significance was determined using an unpaired two-tailed Students t test; *P<0.05.

To determine whether this is of biological relevance, we assessed Dock's capacity to modulate PTP61F-mediated IR regulation in the Drosophila eye. Specifically, as PTP61F had not shown a full suppression of the IR-induced eye overgrowth phenotype (Figure 2, bottom right-hand panel), we asked whether co-expression of Dock could enhance PTP61F's capacity to suppress the IR-driven eye overgrowth. As a control, we also expressed Dock on its own and Dock plus IR. Dock overexpression on its own caused a modest rough eye phenotype (see Supplementary Figure S3 at http://www.BiochemJ.org/bj/439/bj4390151add.htm) and did not suppress the IR-induced eye overgrowth (Figure 4). However, co-expression of Dock plus PTP61F, as compared with PTP61F alone, caused a further reduction in IR-induced eye overgrowth, reducing outgrowth and improving organization (Figure 4). These data are consistent with Dock facilitating PTP61F-mediated down-regulation of IR signalling in vivo.

Figure 4 Dock enhances the PTP61F-mediated suppression of IR-driven eye overgrowth

Control flies (UAS-PTP61F/CyO; UAS-InR/UAS-Dock) and those expressing PTP61F (GMR-gal4/CyO; UAS-PTP61F), IR (GMR-Gal4/CyO; UAS-InR/TM6BTb), IR plus PTP61F (GMR-Gal4/UAS-PTP61F; UAS-InR/TM6BTb), IR plus Dock (GMR-Gal4/UAS-Dock; UAS-InR/TM6BTb) or IR plus PTP61F and Dock (GMR-Gal4/UAS-PTP61F; UAS-InR/UAS-Dock) were processed for scanning electron microscopy (×100) as described in the Materials and methods section. Results shown are representative of two independent experiments (n=4–5 per experiment).

Nck promotes IR dephosphorylation by PTP1B

The Dock protein is highly related to the mammalian adaptor protein Nck that also contains three SH3 domains and one SH2 domain [38,39,50]. In fact, human Nck rescues the dock mutant phenotype of defective photoreceptor cell axon guidance [39]. Moreover, Nck has been implicated in IR signalling [51] and shown to interact via its SH3 domain with IRS-1 [52,53], but it remains unknown if Nck interacts directly with PTP1B and facilitates PTP1B in the attenuation of IR signalling. As a first step, we determined whether Nck and PTP1B could interact in human cells. FLAG–Nck was transiently expressed in HEK-293 cells and its association with endogenous PTP1B and the IR in response to insulin assessed in FLAG or IR immunoprecipitates. Endogenous PTP1B was readily detected in FLAG–Nck immunoprecipitates with and without insulin stimulation (Figure 5A). In contrast, IR association with immunoprecipitated FLAG–Nck was only seen in response to insulin (Figure 5A). Conversely, FLAG–Nck association with the immunoprecipitated IR was increased in response to insulin (Figure 5B). These results indicate that Nck–PTP1B complexes inducibly associate with the Tyr1162/Tyr1163 phosphorylated IR.

Figure 5 Nck complexes with PTP1B and regulates IR signalling

(A) HEK-293 cells expressing FLAG-tagged Nck-1 (Flag-Nck) left untreated or stimulated with 2 nM insulin for 5 min. FLAG immunoprecipitates (IPs) were resolved by SDS/PAGE and processed for immunoblotting as indicated. (B) HEK-293 cells expressing FLAG–Nck were stimulated with 2 nM insulin as indicated and endogenous IR immunoprecipitates were resolved by SDS/PAGE and processed for immunoblotting to detect associated FLAG–Nck. Similar results for (A) and (B) were observed in two independent experiments. (C) HEK-293 cells were transfected with Nck-1-specific or GFP control siRNAs as described in the Materials and methods section to knock down the expression of Nck1. Cells were left untreated or stimulated with 2 nM insulin for 5 min and endogenous IR immunoprecipitates or lysates were resolved by SDS/PAGE and processed for immunoblotting as indicated. (D) Vector control and FLAG–Nck-expressing HEK-293 cells were left untreated or stimulated with 2 nM insulin for 5 min and endogenous IR immunoprecipitates and cell lysates were resolved by SDS/PAGE and processed for immunoblotting to monitor for phosphorylated and activated IR (Phos-IR). (C and D) Densitometric analyses of phosphorylated IR normalized to total IR. Units shown are arbitrary (A.U.) and are means±S.E.M. (n=3); significance was determined using an unpaired two-tailed Students t test; *P<0.05.

Next, we determined whether Nck facilitates PTP1B-mediated IR dephosphorylation. To this end, we first determined whether IR phosphorylation was affected by the transient knockdown of Nck1 in HEK-293 cells using siRNAs. Knockdown of Nck1 by approximately 50% resulted in a significant increase in IR Tyr1162/Tyr1163 phosphorylation in response to insulin (Figure 5C). To complement these results, we then determined if Nck1 overexpression could attenuate insulin signalling. Transient overexpression of FLAG–Nck1 in HEK-293 cells decreased insulin-induced IR Tyr1162/Tyr1163 phosphorylation (Figure 5D). Taken together, these data demonstrate that Nck and PTP1B form a stable complex that inducibly associates with the activated IR for the attenuation of IR Tyr1162/Tyr1163 phosphorylation and signalling.

DISCUSSION

In Drosophila, the functions of the insulin-like growth factor receptor and the IR are served by a single gene, dIR, which controls diverse processes such as cellular growth and proliferation, organ size, lifespan and carbohydrate (trehalose) homoeostasis [4,6]. The IR signalling pathway is highly conserved between mammals and flies, and studies in Drosophila have defined the biological functions of several pathway components, including that of IRS-1–IRS-4, PI3K, Akt, the lipid phosphatase and PTP superfamily member PTEN (phosphatase and tensin homologue deleted on chromosome 10) and the tuberous sclerosis proteins TSC1 and TSC2 [37]. Mutations in genes encoding negative regulators such as PTEN and TSC1/TSC2 result in increased IR signalling and cause hyperproliferative and overgrowth phenotypes, whereas inhibiting IR signalling by mutations, expression of dominant-negatives or the overexpression of negative regulators causes decreases in cell size, cell number and organ size and promotes longevity [3,4,6,5458]. To date, the roles of PTPs in the attenuation of insulin-induced tyrosine phosphorylation-dependent signalling in flies have remained unknown. In the present study we report for the first time that PTP61F is a negative regulator of the Drosophila IR.

Importantly, in the present study, using Drosophila as a model organism, we describe the potential for adaptor proteins to confer PTP substrate selectivity. In particular, we describe the role of the adaptor protein Dock in recruiting and facilitating PTP61F in the dephosphorylation of its substrate, the IR. We found that the expression of Dock promoted IR dephosphorylation, whereas Dock knockdown enhanced IR activation, consistent with Dock and PTP61F synergistically controlling the magnitude of the insulin response. Dock was originally identified as an adaptor for various cell surface receptors, including the IR [41], DASCAM (Down syndrome cell adhesion molecule) [59] and repulsive guidance receptor Robo (Roundabout) [60]. In general, adaptor proteins couple surface receptors to downstream effectors and function to propagate cellular signalling [39]. For example, Dock couples the IR to downstream pathway components in the developing eye and is required for R-cell axon guidance and growth cone morphology [41]. Our results define for the first time the potential for adaptor proteins to not only propagate, but also to terminate tyrosine phosphorylation-dependent cellular signalling by the co-ordinated recruitment of PTPs. Moreover, our results highlight the potential role of adaptor proteins in conferring PTP substrate selectivity. Similar to their PTK counterparts, PTPs exhibit inherent specificity for substrates [8,9]. This is exemplified by catalytic domain swapping experiments between the highly conserved PTP1B and TCPTP, and SHP-1 (SH2 domain-containing protein tyrosine phosphatase 1) and SHP-2 [25,26]. We suggest that adaptor proteins may confer additional specificity by targeting PTPs to defined substrates. It is implicit that PTPs would need to be precisely regulated in such PTP–adaptor protein–substrate complexes. One such mechanism may involve their reversible oxidation and inactivation by reactive oxygen species [6164].

PTP61F was previously identified as a Dock-associated protein [40]. These early studies suggested that the C-terminal region of PTP61F, which contains five proline-rich motifs, may be sufficient for the association and that this may occur via one or more of the N-terminal SH3 domains of Dock [40]. As such, one may expect that the association between PTP61F and Dock would be ligand-independent, as PxxP–SH3 domain interactions are often constitutive. In the present study, we have shown that the PTP61F–Dock complex was constitutive and unperturbed by insulin stimulation. Similarly, PTP1B and Nck formed a stable complex and this too was unchanged by insulin. Although further studies are needed to define the precise residues contributing to the interaction between PTP61F/PTP1B and Dock/Nck, it is likely that the PxxP motifs in PTP61F/PTP1B and the SH3 domains in Dock/Nck play an important role. So how might Dock/Nck target PTP61F/PTP1B to the IR? It has been shown that IR in flies contains a 400-amino acid C-terminal extension, not present in the human IR [65], that contains a PxxP motif [66]. This C-terminal PxxP motif can interact with the Dock SH3 domain in a yeast two-hybrid assay [41]. Other studies have predicted an interaction between tyrosine phosphorylation sites in the Drosophila IR and the C-terminal SH2 domain of Dock [50], whereas genetic studies have suggested that there is a functional redundancy between the SH2 and SH3 domains of Dock during the process of IR-mediated R-cell axon guidance [41]. In the present study, we observed an insulin-induced association between Dock/Nck and tyrosine phosphorylated IR in S2/HEK-293 cells. Accordingly, our studies suggest that the Dock/Nck SH2 domain may play an important role in targeting Dock/Nck to the active IR. This may be particularly pertinent for the mammalian IR, which lacks a PxxP motif. It is also important to point out that the recruitment of Nck/PTP1B to the activated IR may not be direct and instead rely on an intermediate such as IRS-1. This would be in keeping with a previous study that demonstrated that the SH2 domain of Nck-2 can bind to tyrosine phosphorylated IRS-1 in response to insulin [52,53].

Irrespective of the precise molecular basis for the interactions between Dock/Nck, PTP61F/PTP1B and the IR, the present study provides insight into a novel mechanism whereby PTP substrate selectivity may be conferred. In particular, we highlight an important mechanism whereby insulin signalling may be regulated by PTP1B and define the role of the adaptor protein Nck in recruiting PTP1B to the phosphorylated IR. Interestingly, a recent study has shown that the deletion of Nck1 in high-fat fed obese mice improves glucose tolerance and insulin sensitivity [67]. Although the phenotype in Nck−/− mice was ascribed to attenuated ER stress [67], it is also possible that at least part of the phenotype may be attributed to enhanced IR signalling. The highly conserved nature of the PTP active site has presented a significant hurdle in the development of PTP1B-specific inhibitors for use in alleviating insulin resistance in Type 2 diabetes. An exciting alternative approach may be to develop inhibitors that target the interaction between Nck and PTP1B.

AUTHOR CONTRIBUTION

Chia-Lun Wu, Bree Buszard, Coral Warr, Tony Tiganis and Tzu-Ching Meng designed the research. Chia-Lun Wu, Bree Buszard, Chun-Hung Teng and Wei-Lin Chen performed the research. Chia-Lun Wu, Bree Buszard, Chun-Hung Teng, Wei-Lin Chen, Coral Warr, Tony Tiganis and Tzu-Ching Meng analysed the data. Tony Tiganis and Tzu-Ching Meng wrote the paper.

FUNDING

This work was supported by grants from Taiwan's National Science Council [grant number 98-2311-B-001-019-MY3 (to T.-C.M.)], the National Health and Medical Research Council (NHMRC) of Australia (to T.T.) and the Australian Research Council (to C.G.W.). T.T. is a NHMRC Principal Research Fellow.

Abbreviations: Dock, Dreadlocks; dsRNA, double-stranded RNA; EGFR, epidermal growth factor receptor; ER, endoplasmic reticulum; FBS, fetal bovine serum; GFP, green fluorescence protein; HA, haemagglutinin; HEK-293, human embryonic kidney 293; IR, insulin receptor; IRS, insulin receptor substrate; PI3K, phosphatidylinositol 3-kinase; PTK, protein tyrosine kinase; PTP, protein tyrosine phosphatase; RNAi, RNA interference; SH, Src homology; siRNA, small interfering RNA; TCPTP, T-cell protein tyrosine phosphatase; UAS, upstream activating sequence

References

View Abstract