Neuregulin Signaling on Glucose Transport in Muscle Cells*

Neuregulin-1, a growth factor that potentiates myogenesis induces glucose transport through translocation of glucose transporters, in an additive manner to insulin, in muscle cells. In this study, we examined the signaling pathway required for a recombinant active neuregulin-1 isoform (rhHeregulin-β1, 177–244, HRG) to stimulate glucose uptake in L6E9 myotubes. The stimulatory effect of HRG required binding to ErbB3 in L6E9 myotubes. PI3K activity is required for HRG action in both muscle cells and tissue. In L6E9 myotubes, HRG stimulated PKBα, PKBγ, and PKCζ activities. TPCK, an inhibitor of PDK1, abolished both HRG- and insulin-induced glucose transport. To assess whether PKB was necessary for the effects of HRG on glucose uptake, cells were infected with adenoviruses encoding dominant negative mutants of PKBα. Dominant negative PKB reduced PKB activity and insulin-stimulated glucose transport but not HRG-induced glucose transport. In contrast, transduction of L6E9 myotubes with adenoviruses encoding a dominant negative kinase-inactive PKCζ abolished both HRG- and insulin-stimulated glucose uptake. In soleus muscle, HRG induced PKCζ, but not PKB phosphorylation. HRG also stimulated the activity of p70S6K, p38MAPK, and p42/p44MAPK and inhibition of p42/p44MAPK partially repressed HRG action on glucose uptake. HRG did not affect AMPKα1 or AMPKα2 activities. In all, HRG stimulated glucose transport in muscle cells by activation of a pathway that requires PI3K, PDK1, and PKCζ, but not PKB, and that shows cross-talk with the MAPK pathway. The PI3K, PDK1, and PKCζ pathway can be considered as an alternative mechanism, independent of insulin, to induce glucose uptake.

During adult life, NRGs are highly expressed in neurons and they act on the juxtaposed tissues contributing to the maintenance of differentiated patterns, such as the neuromuscular junction, NMJ (8 -12). They are expressed as either membrane or soluble forms, mainly as a consequence of differential splicing of the primary transcript (reviewed in 13 and 14). Soluble forms attach to the basal lamina through proteoglycans (e.g. agrin) and, by regulated proteolysis, they are released and bind to their receptors. NRGs bind to type I tyrosine kinase receptors that belong to the family of EGF receptors, ErbB3 and ErbB4. Binding generates homo-and heterodimerization between them and with ErbB2, which lacks a binding domain, whereas ErbB3 does not have kinase activity (reviewed in Ref. 15). Cultured myotubes express ErbB2 and ErbB3, but not ErbB4 (10,11), so the main NRG receptor is the heterodimer ErbB2/ErbB3. All three receptors are expressed in muscle tissue although only ErbB2 and ErbB4 accumulate at the postsynaptic site (16).
The PI3K/PKB pathway is involved in NRG anti-apoptotic * This work was supported by a grant from the Ministerio de Ciencia y Tecnología, SAF2002-01585 (to A. G.) and grants from the Instituto de Salud Carlos III/FIS, G03/212 and C03/08 (to A. Z.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ Both authors contributed equally to this work. § Supported by the Instituto de Salud Carlos III/FIS, G03/212. ¶ Recipient of a predoctoral fellowship from the Universitat de Barcelona, Spain. § § Supported by the Program Ramón y Cajal from the Ministerio de Ciencia y Tecnología, with FEDER/FSE funds, and the University of Barcelona.
NRGs induce glucose uptake in myocyte cultures and muscle fibers (31). This effect is fast and requires translocation of glucose transporters to the plasma membrane (31). NRGs and insulin induce glucose uptake in an additive manner and both require PI3K activity (31).
A recent study indicates that muscle contraction, caused by either acute exercise or electrical stimulation, induces release of NRGs, to the extracellular milieu that results in ErbBsinduced-activity (40). Both exercise and oxidative stress induce glucose uptake through translocation of glucose transporters, in an additive and independent manner to the action of insulin, and AMPK has been suggested to be involved in this effect (41)(42)(43) although controversy exists in this aspect.
Here we examine the signaling cascade triggering the action of HRG on glucose uptake in L6E9 myotubes. PI3Ks, PDK1, and PKC, but not PKB, are involved in HRG-stimulated glucose uptake. The additive effect of HRG and insulin on glucose uptake did not involve AMPK activation.
2-Deoxy-D-[ 3 H]Glucose Uptake-L6E9 cells were cultured on 6-well plates. 2-Deoxyglucose uptake assays were performed as previously described (31). Radioactivity was determined by scintillation counting. Protein was measured by BCA protein reagent assay (Pierce). Each condition was run in duplicate or triplicate. 2-Deoxyglucose transport was linear during the period (10 min) assayed (not shown).
2-Deoxyglucose uptake assays were performed in strips of rat soleus muscles as previously described (31). Protein was measured by Bradford method.
Preparation of Extracts from L6E9 Myotubes and from Strips of Rat Soleus Muscle-Homogenates were prepared from L6E9 myotubes. They were placed on ice and washed twice in ice-cold phosphate-buffered saline before adding 300 l (6-well plates) or 1 ml (100-mm dishes) of lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% (v/v) Nonidet P-40, 1 mM EDTA, 5 mM sodium pyrophosphate, 1 mM sodium orthovanadate, 50 mM sodium fluoride, containing freshly added protease inhibitors: 0.2 mM phenylmethylsulfonyl fluoride, 1 M leupeptin and 1 M pepstatin). Cells were then scraped, collected into an Eppendorf tube, and then homogenized by repetitive passes through a 25-gauge needle before being centrifuged at 15,000 rpm for 10 min at 4°C. Finally, supernatant was collected and it was stored at Ϫ80°C. Protein was measured by the Bradford method. Preparations of fractions enriched in plasma membranes (PM) or in intracellular membranes (LDM) were obtained from two 10-mm culture dishes as reported (31).
Muscle homogenates were obtained from strips of rat soleus. Muscles were frozen, collected into an Eppendorf, placed on ice containing 200 l of fresh lysis buffer, and then homogenized with an Eppendorf homogenizer. Homogenates were taken to a final volume of 1 ml/strip by adding lysis buffer and then centrifuged at 15,000 rpm for 10 min at 4°C. The pellet was discarded, and supernatant was collected and stored at Ϫ80°C until used. Protein was measured by the Bradford method.
Immunoprecipitation Assays-Immunoprecipitation was performed by conjugating 30 l of protein G-Sepharose beads with 2-5 g of the corresponding antibody, except when it is indicated, shaking for 1 h at 4°C, then washing twice in lysis buffer and incubating with 500 -1,000 g of lysate overnight with constant shaking at 4°C. After brief centrifugation the supernatant was discarded. The pellet was washed several times with the lysis buffer and boiled with 50 l of Laemmli sample buffer (LBS) for Western blot assays as described below. PVDF membranes were blotted with the corresponding antibodies.
For PKCimmunoprecipitation, 500 g of homogenate were incubated overnight with 1 g of an anti-nPKC-antibody (C-20, Santa Cruz Biotechnology), with constant shaking at 4°C, and 30 g of protein G-Sepharose beads were then added. Beads were then washed twice in lysis buffer, boiled at 95°C for 5 min in 50 l LSB, and the supernatant was used for Western blot as described below. PVDF membranes were blotted with anti-phospho-PKC/ (Thr 410/403 ) antibody.
PI3K Activity Assay-L6E9 myotubes were washed once in phosphate-buffered saline and lysed in a buffer containing 10 mM Tris-HCl, pH 7.4, 5 mM EDTA, 50 mM NaCl, 100 mM NaF, and 1% Triton X-100 supplemented with 2 mg/ml aprotinin, 1 mM pepstatin A, 1 ng/ml leupeptin, 2 mM phenylmethylsulfonylfluoride, and 10 mM sodium orthovanadate. 100 g of lysate protein was incubated with 1.5 l of mouse monoclonal anti-phosphotyrosine antibody (PY99, Santa Cruz Biotechnology) for 90 min at 4°C, and then immunoprecipitated with 2.5 mg of rehydrated protein A-Sepharose beads (Sigma) for 60 min. The immunoprecipitates were used for the kinase assay, which was performed with phosphatidylinositol as substrate in the presence of [␥-32 P]ATP (44). Lipids were separated by TLC, and the radioactivity corresponding to phosphatidylinositol 3-phosphate was analyzed using a Fuji FLA-2000 phosphoimager.
PKB, p70S6K, p38MAPK, and p42/p44MAPK Activity Assays-5 g of sheep antibodies raised against PKB␣ or ␥ (45), p70S6K (46), MAP-KAP-K2 (47), or p90rsk (48) were conjugated to 5 l of protein G-Sepharose beads for 30 min at 4°C. After washing twice in lysis buffer, 500 g of total lysate in the case of p70S6K or 100 g otherwise were added and the mixture was incubated for 1 h at 4°C on a shaking platform. It was then centrifuged briefly at 13,000 rpm. The immunoprecipitates were washed twice with 1 ml of lysis buffer containing 0.5 M NaCl, and twice with 1 ml of reaction buffer (50 mM Tris-HCl, pH 7.5, 0.1% (v/v) ␤-mercaptoethanol, 0.1 mM EGTA). The standard assay, which is employed unless stated, otherwise contained (50 l of total volume): washed protein G-Sepharose immunoprecipitated, 50 mM Tris-HCl, pH 7.5, 0.1 mM EGTA, 0.1% (v/v) ␤-mercaptoethanol, 2.5 M PKI (TTYADFIASGRTGRRNAIHD, peptide inhibitor of the cyclic AMP-dependent protein kinase), 10 mM magnesium acetate, 0.1 mM [␥-32 P]ATP (ϳ300 cpm/pmol) and the corresponding substrate: 30 M Crosstide peptide (GRPRTSSFAEG) in all cases except for measuring MAP-KAP-K2 activity (30 M MAPKAP-K2 substrate KKLNRTLSVA). The assays were carried out for 30 min at 30°C, the assay tubes being agitated continuously to keep the immunoprecipitated in suspension. Incorporation of [␥-32 P]phosphate into each peptide substrate was determined using p81 phosphocellulose paper. Papers were washed in 0.5% orthophosphoric acid, dried, and Cherenkov radiation was counted. 1 milliunit of activity is the amount of enzyme that catalyzes the phosphorylation of 1 pmol of substrate in 1 min.
AMPK Activity Assay-AMPK ␣1 and ␣2 activities were measured in L6E9 myotubes lysates as described previously (48). Briefly, cells were incubated in HEPES-buffered saline (HBS) containing 5 mM glucose for 30 min at 37°C. After washing the cells in phosphate-buffered saline, they were lysed into a minimal volume of buffer A (50 mM Tris-HCl, pH 7.5, 1 M EDTA, 1 mM dithiothreitol, 10% glycerol, 50 mM sodium fluoride, 5 mM sodium pyrophosphate, 1 mM benzamidine, 0.1 mM phenylmethylsulfonyl fluoride, and 1% Triton X-100), cell debris was spun down in a bench top centrifuge at 12,000 rpm for 5 min, and the supernatant was taken for assay. Protein concentration was determined using the Bradford reagent. 100 -200 g of protein was then sequentially immunoprecipitated using sheep antibodies raised against either ␣1 or ␣2 (49) conjugated to protein G-Sepharose beads. AMPK activity within the immune complex was measured by the phosphorylation of synthetic SAMS peptide using radiolabeled [␥-32 P]ATP.
Immunoblotting-Protein samples containing LSB were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to PVDF membranes as previously described (31). The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline solution (TBS) for 1 h at room temperature and then incubated overnight with the corresponding primary antibodies at recommended dilutions in 1% bovine serum albumin. PVDF membranes were then washed, incubated with appropriate secondary antibodies, containing horseradish peroxidase, for 1 h at room temperature and washed again. Proteins were detected by the ECL method and quantified by scanning densitometry.
Statistical Analysis-Data are presented as means ϩ S.E. Unpaired Student's t test was used to compare two groups and one-way analysis of variance, posthoc Duncan's t test, was used to compare more than two groups.

Neuregulins-induced Glucose Transport Requires an Intact
ErbB3 Binding Activity-In order to prove the necessity of the ErbB3 binding activity in the NRG action on glucose uptake in L6E9 myotubes, we used an antibody, Ab5, that binds to ErbB3 and blocks the NRG-binding domain. When Ab5 (10 g/ml) was added to the medium 30 min before HRG, it abolished the stimulation of glucose transport (Fig. 1A). Ab5 did not alter the insulin effect on glucose uptake (Fig. 1A).
HRG enhanced tyrosine phosphorylation of ErbB3 receptors but it did not induce tyrosine phosphorylation of the ␤ subunit of insulin receptors (Fig. 1B). Moreover, chronic insulin treatment that generated insulin resistance characterized by a deficient induction of glucose uptake, did not alter HRG action (Fig. 1, C and D). In all, HRG required ErbB3 to stimulate glucose uptake in L6E9 myotubes and this effect did not involve activation of the insulin receptor and it was not modified by induction of insulin resistance.
HRG (3 nM) activated PKC in L6E9 myotubes, with a maximum at 30 min (Fig. 2C). HRG and insulin did not show additive effects (Fig 2, A-C). PDK1 Is Involved in the Effect of Neuregulins on Glucose Uptake in L6E9 Myotubes-PI3K activity is required in the HRG-induced pathway on glucose uptake in L6E9 myotubes, based in our previous studies using the specific PI3K inhibitor wortmannin (31). In the present study, we analyzed the role of PDK1 on the NRGs effects, based on the use of TPCK, a PDK1 inhibitor (52). 50 M TPCK inhibited both PKB and PKC activities induced by HRG or by insulin (Fig. 3, A and B). TPCK inhibited PKB Thr 308 phosphorylation without affecting Ser 473 phosphorylation (Fig. 3A). Interestingly, Ser 473 was consistently phosphorylated less under HRG action than under insulin action while no remarkable differences were observed at Thr 308 phosphorylation. 50 M TPCK also repressed both HRGand insulin-induced glucose transport (Fig. 3C).
PKB Is Not Involved in the Action of Neuregulins on Glucose Uptake in L6E9 Myotubes-In order to study the role of PKB in the HRG action on glucose uptake, we used dominant negative mutants of PKB. Initially, we transduced L6E9 myocytes with adenovirus containing mutated forms of PKB␣, AA-PKB, and AAA-PKB (MOI 300, 48 h), in order to inhibit endogenous PKB activity. L6E9 myocytes transduced with double mutant viruses showed a partial inhibition of insulin-or HRG-induced PKB activity (Fig. 4A). Almost complete inhibition of the PKB activity was obtained with the triple mutant AAA-PKB (Fig.  4A). The overexpression of the double mutant AA-PKB induced a significant inhibition of insulin-stimulated glucose transport, which was even greater in cells expressing the AAA-PKB mutant (Fig. 4B). In these conditions the dominant negative forms of PKB did not inhibit HRG-induced glucose transport or HRGinduced GLUT4 translocation (Fig. 4, B and C).
PKC Activity Is Necessary for Neuregulin-induced Glucose Uptake-The next step was to determine the involvement of PKC in the HRG-induced glucose uptake. Transduction of L6E9 myotubes with KI-PKC viruses (MOI 100, 48 h) blocked PKC activity (Fig. 5A) and repressed the effects of HRG and insulin on glucose uptake (Fig. 5B).
P70S6K, p38MAPK, and p42/p44MAPK Are Not Directly Involved in the Action of Neuregulin on Glucose Uptake-We analyzed HRG action on p70S6K, p38MAPK, or p42/ p44MAPKs activities in L6E9 myotubes and their possible involvement in the effects of HRG on glucose uptake using specific inhibitors.
HRG induced p70S6K activity (Fig. 6A), which was completely abolished by the specific inhibitor rapamycin. However, rapamycin did not alter the effect of HRG on glucose uptake (Fig. 6B). p38MAPK was moderately activated by HRG in comparison with the hyperosmotic effects of 0.5 M sorbitol (Fig. 6A). The specific inhibitor SB203580 (10 M) abolished HRG activation of p38MAPK activation by HRG action (Fig. 6A) but did not interfere with its action on glucose uptake (Fig. 6B).
Effects of Neuregulins on AMPK Activity-Due to the lack of a specific inhibitor of AMPK, many studies have used 5-amino-4-imidazolecarboxamide riboside, AICAR, as an AMPK activator. However, AICAR does not lead to activation in many cell types since its efficacy is dependent on its uptake by the cell and accumulation as a monophosphorylated form, ZMP, which acts as an AMP-mimetic, promoting allosteric activation of AMPK and phosphorylation by an upstream kinase (43). L6E9 myotubes are not a good cellular model for AICAR trials since it appears to have low effect on AMPK ␣ 1 and ␣ 2 (not shown), so AMPK was activated by increasing intracellular AMP with DNP, an oxidative phosphorylation uncoupler. DNP stimulated AMPK␣ 1 and AMPK␣ 2 activities in L6E9 myotubes. Under these conditions, HRG did not alter the AMPK activity at any   FIG. 2. HRG stimulates PI3K, PKB␣/␥, and PKC. L6E9 myotubes were treated with 3 nM HRG or/and 1 M insulin for the indicated times, and each kinase was immunoprecipitated from the lysates and assayed. A, PI3K activity was assayed and phosphatidylinositol 3-phosphate (PIP 3 ) production was quantified densitometrically. Results are relative values that correspond to a representative experiment from two independent experiments, each one run in duplicate and assayed twice. B, PKB␣ or PKB␥ were immunoprecipitated, and activities were assayed as described under "Experimental Procedures." Results correspond to a representative experiment from two different experiments, each one run in duplicate and assayed in triplicates. C, total PKC was immunoprecipitated from lysates and phospho-Thr 410 -PKC was immunodetected and quantified by densitometric analysis. Results, expressed as relative to basal values, were obtained from three different experiments. B, basal; H, HRG, I, insulin. * indicates significant differences with the basal group, at p Ͻ 0.01. time or dose studied (Fig. 7A). Identical results were obtained in the H-2K b myocytes (48) (not shown).
DNP, HRG, and insulin showed additive effects on glucose transport (Fig. 7B) indicating that they activate at least partially different intracellular pathways to induce glucose transport in muscle cells.
Neuregulins Signaling Cascade on Glucose Transport in Rat Soleus Muscle-Strips of rat soleus muscle were incubated with HRG (5 nM, 15 min) or insulin (100 nM, 10 min) and lysates were obtained, immunoprecipitated with an anti-phosphotyrosine antibody, and the pellets were immunoblotted with antibodies against each neuregulin receptor. Results indicated that HRG, but not insulin, activated all the expressed neuregulin receptors in muscle fiber, ErbB2, ErbB3, and ErbB4 (Fig. 8A).
Treatment with wortmannin (2 M, 150 min) abolished both insulin and HRG action on glucose uptake in incubated strips of rat soleus (Fig. 8B), indicating that PI3K is also required in the neuregulins signaling cascade at the muscle fiber.
Next, strips of rat soleus muscle were treated with HRG or insulin during 30 min, and lysates were obtained to analyze phosphorylation levels of PKB and PKC. Whereas insulin induced phosphorylation of both kinases, HRG was unable to induce PKB phosphorylation (Fig. 8, C and D) at any time ranging from 10 to 60 min (not shown).

DISCUSSION
In this study we have demonstrated that NRGs stimulate PI3K, PKB, PKC, p70S6K, p38MAPK, and p42/p44MAPKs in L6E9 myotubes, that the pathway PI3K, PDK1, and PKC is essential to induce glucose transport and that p42/ p44MAPKs also contribute to a maximal glucose transport stimulation. Studies done in rat soleus muscles also support the implication of the PI3K pathway on NRG-stimulated glucose transport. This is the first report indicating that NRGs induce a PDK1/PKC pathway. In L6E9 myotubes ErbB3 is required to initiate this signaling cascade. NRGs do not transactivate the insulin receptors, so the NRG signaling pathway can be considered as an alternative mechanism to induce glucose transport in muscle cells which, like insulin, is initiated by PI3K activation (31,53,54). ErbB3 contains six consensus sequences, YXXM motifs that bind to the SH 2 domain of the PI3K p85 subunit after ligand-induced tyrosine phosphorylation (19). ErbB3 has been characterized as the main mediator of heregulin-dependent PI3K activation pathway (55) and expression of mutated forms of ErbB3 at the YXXM motifs in COS7 cells abolishes PI3K and PKB activation by heregulin binding to the heterodimer ErbB2/ErbB3 (56). Since ErbB4 is also expressed and activated by HRG in skeletal muscle, differences on the insulin and the HRG action in soleus and cultured myocytes could be consequence of a different pattern of receptors expression.
PI3K activity is essential for HRG action on glucose transport (31), although HRG induces lower maximal PI3K activities than insulin. Differences on PI3K activation could be a consequence of the expression levels of ErbB3 and insulin receptors in L6E9 myotubes, and they also suggest that PI3K activity might not be limiting for signaling on glucose uptake. This view is also supported by the lack of additivity of HRG and insulin on PI3K activity. Otherwise, HRG requires more time than insulin to reach maximal effects on glucose transport (31), so a critical amount of phosphatidylinositol 3-phosmin). Then, the 2-deoxyglucose uptake assay was performed. Results are obtained from three different experiments. Basal uptake: 0.62 Ϯ 0.03 pmol/g of protein ϫ 10 min. B, basal; I, insulin action; H, HRG action. * indicates significant differences with the control group, at p Ͻ 0.001.

FIG. 3. HRG-induced glucose uptake requires PDK1 activity.
L6E9 myotubes were treated with 3 nM HRG (15 min) or 1 M insulin (10 min), in the absence or presence of the PDK1 inhibitor, 50 M TPCK (35 min). A, total PKB activity was determined from one experiment assayed in triplicate. Cell lysates were also immunoblotted with antibodies that recognize total PKB (C-terminal) or PKB phosphorylated at Ser 473 (hydrophobic motif) or Thr 308 (activation loop). Similar results were obtained in three separate experiments. B, phospho-Thr 410 -PKC was immunodetected and densitometrically quantified. Relative phospho-Thr 410 -PKC levels are the mean of three independent experiments. C, L6E9 myotubes were treated with 3 nM HRG (90 min) or 1 M insulin (30 min), in the absence or the presence of 50 M TPCK (110 phate may be needed to trigger HRG stimulation of glucose transport. It is notable that HRG action is not impaired in muscle cells made insulin resistant by chronic exposure to insulin. Interestingly, soleus muscle insulin responsiveness was reduced in the adult rat compared with 1-month-old rats (2-3-fold increase in response to maximal insulin in adult rats and 5-9-fold increase in young rats, respectively); under these conditions, heregulininduced stimulation of glucose transport remained unaltered (1.5-2-fold increase both in adult and in young rats). 2  studies are being addressed to investigate the extent of neuregulin action in muscle in animal models of type 2 diabetes, obesity or aging.
HRG induces both PKB␣ and PKB␥ in L6E9 myotubes. Similar results were previously reported for insulin action in L6 myotubes (45). HRG induces lower PKB␣ activity than insulin, as for PI3K activity. This is in accordance with the lower ability of HRG to phosphorylate Ser 473 , which is involved in the activation of PKB. In contrast, HRG has stronger maximal effects than insulin on PKC, but HRG requires longer time than insulin to reach them (maximal insulin effect on PKC is reached within 10 min, not shown). These differences might explain the potentiated effects observed on glucose transport since insulin and neuregulins do not have additive effects on PKB or PKC activation levels. Nonetheless, an alternative pathway to PI3K activation cannot be ruled out. In fact, a second insulin pathway is involved in the stimulation of glucose transport in adipose cells (reviewed in Ref. 57). This second pathway requires phosphorylation of the adaptor protein Cbl by the insulin receptor and activation of TC10, a small GTP-binding protein located in lipid rafts, which modulates actin structure. At present we do not know whether neuregulin stimulates the Cbl pathway in muscle cells and, if so, whether it is involved in neuregulin-induced glucose uptake.
Several lines of evidence implicate PKB activity in insulininduced glucose uptake both in adipocytes and muscle (32)(33)(34)(35)(36) although some controversy exists (58,59). PKB␤ has been implicated in insulin action on glucose transport in 3T3-L1 adipocytes (36). Our results indicate that PKB blockage alters only partially insulin action on glucose uptake, suggesting that PKB might not be essential in L6E9 myotubes. Moreover, there is no dependence on PKB activity in the NRGs action on glucose transport in muscle cells and HRG does not stimulate PKB phosphorylation in soleus. Then, PKB does not seem to be required in the pathway inducing glucose uptake in response to NRGs. One interesting possibility that should be explored is that a different subcellular localization of PI3K is responsible for the different downstream activities of insulin or neuregulin that trigger glucose transport.
The effect of HRG on glucose transport is also independent of AMPK activity. DNP induces glucose uptake in an additive manner to NRGs action, so AMPK might constitute another pathway, independent of insulin and NRGs, to activate glucose uptake. Both insulin and exercise induce the MAPK pathway, in a PI3K-dependent manner and an AMPK-dependent manner, respectively (60,61). Whereas PD98059 treatment does not affect insulin action on glucose uptake in muscle cells (62), it impairs AICAR stimulation of glucose transport in EDL muscle (61). Thus, the partial impairment of HRG action on glucose uptake under PD98058 treatment suggests that the PI3K-PDK1-PKB pathway activated by NRGs is modulated by the MAPK pathway. We conclude that NRGs require a signaling pathway to induce glucose uptake in L6E9 myotubes that involves ErbB3, PI3K, PDK1, and PKC and is not dependent on PKB activity. A, strips were incubated with insulin (100 nM, 10 min) or HRG (5 nM, 15 min). Lysates were obtained and immunoprecipitated with anti-phosphotyrosine antibody. Pellets were immunoblotted for ErbB2, ErbB3 and ErbB4. Results are shown as representative autoradiograms of three different samples. B, strips were incubated with insulin (100 nM, 60 min) or HRG (5 nM, 120 min), in the absence or presence of wort-