|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DEVELOPMENTAL PHYSIOLOGY AND PREGNANCY
Departament de Fisiologia, Facultat de Biologia, Universitat de Barcelona, Barcelona, Spain
Submitted 26 April 2005 ; accepted in final form 25 January 2006
| ABSTRACT |
|---|
|
|
|---|
insulin; mitogen-activated protein kinase; Akt
These various stimulatory actions of IGF-I and insulin are ultimately dependent upon peptides binding to their respective cell surface receptors. Both receptors are heterotetrameric, consisting of two extra cellular
-subunits and two
-subunits crossing the cell membrane, which possesses a tyrosine kinase activity (TKA) domain within its intracellular area (28, 48). In mammalian muscle, insulin receptors predominate over IGF-I receptors, while fish skeletal muscle exhibits higher levels of IGF-I receptors than insulin receptors (36, 37). On the basis of these findings, we have recently shown the abundance of IGF-I receptors in trout skeletal muscle myosatellite cells during periods of low insulin binding (8). Taken together, these data suggested an important role for IGF-I in fish skeletal muscle. This was confirmed more recently when IGF-I was discovered to stimulate glucose and alanine uptake in trout muscle cells to a greater extent than insulin. Furthermore, IGF-I was found to stimulate cell proliferation in this model, whereas insulin did not (7).
Different studies have examined insulin and IGF-I receptor TKA in both mammalian (20, 53) and fish skeletal muscle (37). Indeed, the authors addressed the entire fish life cycle (33) and showed that IGF-I receptors exhibit higher TKA than insulin receptors, a result that is contrary to that described in mammalian muscle (where insulin receptors have higher TKA than IGF-I receptors); moreover, fish muscle exhibits lower TKA levels than mammalian muscle (37). These differences are undoubtedly related to the potentially different roles played by insulin and IGF-I in fish skeletal muscle (7, 37, 39).
Cross-talk between these growth-related hormones is widespread: IGF-I and insulin converge via common message transduction systems, including the phosphatidylinositol 3-kinase (PI3-kinase) and MAPK cascades, while IGF-I and insulin can potentially bind to each others receptors. Mammalian studies have shown that following receptor auto-phosphorylation in tyrosine residues, the signal is transmitted downstream (50), thereby phosphorylating other cell proteins in turn. (reviewed in Refs. 35 and 46). For skeletal muscle insulin and IGF-I receptors, this signal may be transmitted by two main methods: the PI3-kinase-Akt and MAPK pathways. IGF-I is able to stimulate both proliferation and differentiation of muscle cells in culture (12, 13, 15, 43). Proliferation as a response to IGF-I receptor (IGF-IR) signaling is mainly mediated through the activation of the MAPK pathway, which has been implicated in increasing muscle cell proliferation (9, 10, 21, 25, 29), whereas signaling through the PI3-kinase-Akt pathway mediates such crucial effects in skeletal muscle as myoblast differentiation and anabolic effects, including glucose and amino acid uptake (6, 9, 15, 22). Active Akt induces mammalian myotube hypertrophy (42), while in myotubes, but not in myoblasts, Akt is able to inhibit MAPK activation (42, 52).
The MAPK and PI3-kinase-Akt pathways exert completely different effects in mammalian skeletal muscle, and the mechanism by which myoblasts switch from a proliferative to a differentiative signal still remains unclear [reviewed by LeRoith (27)].
In fish, very little is known about the IGF-I and insulin signaling pathways. Although Pozios et al. (40) showed that IGF-I stimulates proliferation of zebrafish embryonic cells by activating the MAPK and PI3-kinase signaling pathways, no studies on IGF-I or insulin intracellular signaling in fish skeletal muscle have been performed thus far, mainly because of the lack of an appropriate in vitro model of fish muscle growth and development. In this respect, the development of a technique for the primary culture of rainbow trout skeletal muscle cells (14, 41) has opened up possibilities for studies on insulin and IGF-I actions in vitro, as we have previously conducted (7, 8).
The present study, which employed a chromatographic technique to separate insulin and IGF-I receptors, as well as a primary culture of trout skeletal muscle cells, had several aims: 1) to compare the intrinsic TKA of the insulin and IGF-I receptors in fish, 2) to identify components of the intracellular signaling cascades of insulin and IGF-I in trout skeletal muscle cells, and 3) to determine their activation in response to both peptides throughout the in vitro differentiation of skeletal muscle cells.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Animals and cell cultures.
Rainbow trout (Oncorhynchus mykiss) experimental animals were obtained from the Piscifactoria Truites del Segre (Oliana, Lleida, Spain) and maintained in the Servei dEstabulari of the Faculty of Biology, University of Barcelona, as described previously (7). The Animal Care and Use Committee of the University of Barcelona reviewed and approved the treatment protocol for all animal experiments. For IGF-I and insulin receptor signal transduction experiments, primary cultures of rainbow trout skeletal muscle cells were carried out: for each culture, 30 to 60 animals, each weighting
5 g, were killed by a blow to the head and then immersed for 30 s in 70% ethanol to sterilize external surfaces. Cells were isolated, pooled, and cultured following previously described protocols (8, 41). All experiments were conducted with cells seeded at a density of 1.5 to 2 x 106 per well, in six-well plastic plates (9.6 cm2/well, Nunc, Roskilde, Denmark). Observations of morphology were made regularly to control the cell state on day 1 (mononucleated cells), day 4 (mostly small myotubes), and day 11 (big myotubes). All experiments were performed in triplicate, as was each condition (three wells). Cells were incubated at 18°C, the optimal temperature for culture growth.
For IGF-I and insulin receptor experiments, brown trout larvae (Salmo trutta) obtained from the Piscifactoria de Bagà (Barcelona, Spain) were used. Samples were frozen directly in liquid nitrogen, where they were kept until the insulin and IGF-I receptors were purified.
Separation of IGF-I and insulin receptors. The first step in separating insulin and IGF-I receptors by differential centrifugation was conducted in S. trutta larvae following the protocol described by Maestro et al. (31). Frozen samples (7 g) were homogenized in a Tris·HCl buffer (25 mM Tris·HCl, 5 mM CaCl2, pH 7.6) and centrifuged at 600 g for 10 min at 4°C. Supernatant was taken and centrifuged again at 40,000 g for 30 min at 4°C. Both pellets were resuspended in a buffer containing (in mM): 25 HEPES, 4 EDTA, 4 EGTA, 2 PMSF, 1 bacitracin, 1 leupeptin, 1 pepstatin, and 25 benzamidine, pH 7.6. Membrane solubilization was performed by adding Triton X-100 to a final concentration of 2%, which was maintained in agitation for 1 h at 4°C. Subsequently, the sample was centrifuged at 150,000 g for 90 min at 4°C, and supernatant was passed three times through a WGA-agarose column. After we washed the column with buffer (25 mM HEPES and 0.1% Triton X-100, pH 7.6), receptors were eluted with washing buffer supplemented by N-acetyl-D-glucosamine (0.3 N).
Both the insulin and IGF-I receptors were purified following the methods described by Yu et al. (51) and Tollefsen et al. (49) with certain modifications. This technique is based on the coupling of a cold peptide (insulin or IGF-I) to an agarose gel (Affigel 10, Bio-Rad), thereby obtaining either an insulin-agarose or an IGF-I-agarose column. Briefly, to obtain the latter, agarose gel was incubated with 30 µg of IGF-I in a coupling buffer (0.1 M HEPES, 6 M urea, 80 M CaCl2, pH 7.6) for 4 h at 4°C. For insulin-agarose, incubation was performed using 45 µg of cold insulin, and under the same time and temperature conditions. The coupling reaction was halted by adding ethanolamine 1 M at pH 8.0. The resulting columns were equilibrated with C buffer (0.1 M HEPES, 6 M urea, pH 7.6) and were then ready for use. To separate IGF-IR, samples (S. trutta larvae) semipurified in a WGA-agarose column were diluted with C buffer and incubated with 1 µl of insulin 350 µM to saturate the insulin receptor (INS-R) for 30 min at 22°C (this step was not necessary for insulin-agarose). Afterward, the sample was saturated with a solution of IGF-I-agarose (in a 12:1 proportion, sample: IGF-I-agarose), and the entire complex was incubated for 16 h at 4°C with agitation. The mixture was then passed through a column, where the IGF-I bound to the agarose-fixed IGF-I receptors. After eluting the sample with buffer (10 mM sodium acetate, 1.5 M NaCl, 0.1% Triton X-100, 10% glycerol, pH 5.0) and neutralizing it with Tris·HCl, pH 7.4, several fractions containing purified IGF-IR were obtained. By substituting cold IGF-I with cold insulin, the steps followed for INS-R purification were identical.
Western blot analysis. To determine the intracellular signal transduction pathways activated by insulin and IGF-I during in vitro development of skeletal muscle cells in culture, Western blots against several proteins involved in the PI3-kinase-Akt and MAPK pathways were performed.
In these experiments, 3050 fish, rainbow trout (Oncorhynchus mykiss) each weighting
5 g, were used for every culture. After pooling cells from all animals of the same culture, the cells were seeded at a density of 1.5 to 2 x 106 per well in six-well plastic plates. After 1, 4, or 11 days in culture, the cells were incubated with DMEM + 0.5% BSA for 23 h. Afterward, they were incubated with DMEM+ 0.5% BSA and a fixed concentration of peptides (100 nM for IGF-I or 1 µM for insulin) for 30 min. Subsequently, the medium was aspirated, the wells were washed with ice-cold PBS, and the cells were lysed with lysis buffer (1% Nonidet P-40, 0.4 mM sodium orthovanadate, 10 mM Tris, 140 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM PMSF, 10 µg/ml aprotinin, 10 µg/ml leupeptin, pH 7.6). After obtaining cell lysates, we determined protein content using the Bradford reagent method (3) and using a polyacrylamide gel, we performed electrophoresis in the presence of SDS (SDS-PAGE) (each lane was loaded with 50 µg of protein). Samples were then transferred to a PVDF membrane for 90 min under a constant current of 1A in a transfer buffer (25 mM Tris·HCl, 190 mM glycine, 20% methanol, pH 7.5). After a 30-min wash (10 mM Tris·HCl, 100 mM NaCl, 0.1% Tween 20, pH 7.5), the membrane was incubated overnight at 4°C with the primary antibody (diluted in a washing buffer, at the dilution indicated at the bottom of each figure). The primary antibodies detected the presence of MAPK and its active form MAPK-P, as well as the Akt protein and its phosphorylated form (Akt-P).
The membrane was then washed for 30 min and incubated for 1 h at room temperature with the corresponding secondary antibody at the indicated dilution. Immunoreactive bands were visualized by ECL and quantified with an image analyzer (TotalLab v1.00, Nonlinear Dynamics, 2000).
To better characterize the signal transduction pathways, we analyzed the effects of two inhibitors of the cell signaling: the PD-98059 is an inhibitor of the MEK1 protein, a component of the MAPK pathway, whereas wortmannin is an inhibitor of the PI3-kinase-Akt pathway. Cells were preincubated for 30 min with wortmannin (1 µM) or PD-98059 (50 µM), and IGF-I was added for 30 additional minutes.
Binding assays in purified receptor preparations.
Binding experiments were performed in purified receptor preparations from S. trutta larvae following the method described by James et al. (20). A 50-µl eluted sample from an insulin-agarose column and a 40-µl eluted sample from an IGF-I-agarose column (5- to 10-fold concentrations by centrifugation at 5,000 g for 15 min at 4°C using Filtron 30K Microsep Centrifugal Concentrators), corresponding to
3 µg of INS-R and 3 µg of IGF-IR, respectively (Bradford method; Bio-Rad), were incubated for 16 h at 4°C in a buffer of 30 mM HEPES, including 0.1% BSA and 100 units/ml of bacitracin (pH 7.6), with increasing concentrations of nonlabeled hormone (ranging from 0.0125 nM to 100 nM porcine insulin or recombinant human IGF-I) and the radioactive ligand in trace concentrations (25 pM). Receptors that were bound to the ligand were precipitated with bovine gamma globulin (0.08%) and polyethylene glycol (10.4% wt/vol), and then centrifuged at 14,000 g for 7 min at 4°C. Nonspecific binding was calculated as follows: a radioactive-bound ligand, in the presence of cold hormone at a concentration of 100 nM, was subtracted from the total count.
TKA assays.
TKA was determined in preparations of purified receptors from S. trutta larvae according to the method described by James et al. (20). Purified receptors (34 µg of protein) were incubated for 16 h at 4°C with cold insulin or IGF-I (60 nM, final concentration) in a HEPES buffer in the presence of 100 mM MgCl2, pH 7.4. Samples were later incubated with 50 µM [32
P]ATP for 10 min to allow receptor auto-phosphorylation. A synthetic substrate poly (Glu/Tyr; 4:1) was added at a final concentration of 0.25 mg/ml and incubated for 30 min. The reaction was halted by transferring the samples to filter paper (Whatmann 3MM) and washed in 10% trichloroacetic acid with 10 mM sodium pyrophosphate. Radioactivity was recorded on filter paper using a scintillation counter.
Statistical analysis. The treatments were performed in triplicate for each experiment, and each experiment was performed three times. Data are presented as means ± SE. For the receptor purification experiments, at a minimum, each binding experiment was performed in duplicate using receptors from different purifications. Statistical differences between conditions were tested with one-way ANOVA, followed by the Tukey test. Results were considered statistically significant at P < 0.05
| RESULTS |
|---|
|
|
|---|
In samples eluted from the insulin-agarose column (preparations containing a high proportion of INS-R), insulin stimulated phosphorylation of the exogenous substrate 122.43% above the basal level. In preparations obtained from the IGF-I-agarose column (containing a high percentage of IGF-IR), IGF-I stimulated TKA 162.98% above the basal level (Table 1). In this table, we also record the binding observed in these preparations for each peptide and the TKA/binding ratio. Results show that insulin receptors exhibited a threefold greater activity than IGF-I receptors.
|
Insulin and IGF-I signaling pathways in skeletal muscle. To study the activation of different IGF-I and insulin receptor signal transduction pathways, we determined the presence of both total and phosphorylated MAPK and Akt proteins throughout the different developmental stages of the trout skeletal muscle cells in a primary culture.
MAPK pathway. These Western blots were performed from lysates of rainbow trout skeletal muscle cells in primary cultures at different stages of development: muscle cells on day 1 of culture development, myoblasts (small myotubes formed by fusion of few cells) on day 4, and large myotubes on day 11. In Fig. 1, we show a representative experiment, as well as a densitometric analysis of the results; throughout development, from day 1 to day 11, total MAPK levels increased in control cells (Fig. 1A). In addition, both IGF-I and insulin caused a significant increase in total MAPK levels compared with the control sample on day 1 (nondifferentiated cells), whereas on day 4 (small myotubes) and in fully differentiated myotubes (day 11), the insulin effects were not significant (Fig. 1B). When treating myoblasts with the protein synthesis inhibitor cycloheximide, the effect on MAPK total levels caused by IGF-I was not observed (data not shown).
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
TKA of IGF-I and insulin receptors. Here, we describe the phosphorylation capacity of insulin and IGF-I receptors in trout and, therefore, their potential to develop intracellular signaling. To study TKA in IGF-I and insulin-purified receptors, we used brown trout larvae, as our group had previously described the stages of development in brown trout, in which the binding of these peptides is highest (33). Our aim was to compare the intrinsic TKA of both receptors to establish which possesses the greater phosphorylative capacity, and thereby identify the higher transmission potential of the intracellular signal and the resulting effects generated by that receptor. When comparing the TKA of these receptors we observed that although the IGF-I binding levels in the purified IGF-IR preparations were much higher (almost fourfold) than the purified INS-R insulin preparations, IGF-I stimulation of TKA was only slightly higher, although not significantly so than that for insulin (163% vs. 122%, respectively). In addition, when calculating the ratio between TKA and binding in the purified receptor preparations, we observed INS-R had a ratio nearly four times greater than that for IGF-IR (24.73 vs. 7.31), indicating a higher phosphorylative capacity for INS-R compared with IGF-IR. These results confirm previous studies hypotheses, as they are consistent with data obtained in several trout tissues (26), as well as in other poikilotherms (37). It has, moreover, been suggested that through evolution, INS-R has acquired a higher phosphorylative capacity in comparison to mammals and fish (39).
Signal transduction in skeletal muscle. In mammals, the PI3-kinase and MAPK pathways are the main signaling pathways for IGF-I and insulin (4, 28). It has been postulated that such pathways exert opposite effects in muscle, with the respective consequences dependent on the myocyte differentiation state. Cross-talk can also exist, however, particularly at the Ras protein level, acting as a positive effector in both pathways (19, 23).
The primary culture of trout skeletal muscle cells has been extensively shown as a very powerful in vitro tool to study the IGF-I and insulins role in fish skeletal muscle (7, 8). Preliminary experiments identified the presence of these IGF-I and insulin signaling pathway components in these cells; PI3-kinase, Akt, and MAPK, as well as their active forms, are all present in trout skeletal muscle cells, which can be observed when the cells are activated with either IGF-I or insulin (data not shown). To determine the activation of these proteins throughout muscle development, cells at different stages of in vitro development were utilized, from myoblasts to fully differentiated myotubes. The incubation time for IGF-I or insulin in fish cells (30 min) has been shown to be sufficient for MAPK and Akt protein phosphorylation in a fish model, as described by Pozios et al. (40) using zebrafish embryo cells. The specificity of the detected molecules was proved through the use of specific inhibitors of the MAPK and PI3K-Akt signaling pathways: when incubating trout skeletal muscle cells with the inhibitor wortmannin, we observed a decrease on Akt phosphorylation, as it has been previously described (44, 47); on the other hand, the incubation of skeletal muscle cells with the inhibitor PD-98059 caused a decrease on the IGF-I-stimulated MAPK phosphorylation, in agreement with the results obtained by Pozios et al. (40) on zebrafish cells.
In the cultured cells, we found that IGF-I and insulin caused an increase in total MAPK levels, particularly on day 1 (myoblasts) and that this effect decreased as the culture developed into myotubes by day 11. Experiments with the protein synthesis inhibitor cycloheximide (where a preincubation of cells with the inhibitor provoked a reduction of the IGF-I-stimulated increase of total MAPK levels) suggest that this increase of MAPK immunoreactivity levels could be linked to a synthesis of new protein, a hypothesis that should be confirmed in future studies. In addition, both IGF-I and insulin increased MAPK-P levels above the control in both myoblasts (day 1) and small myotubes (day 4), whereas there was no significant change in differentiated myotubes. Taken together, these results suggest that IGF-I and insulin activate the MAPK pathway in myoblasts but not in differentiated muscle cells. That is, activation of the MAPK pathway in trout skeletal muscle cells is stage-dependent, an observation supported by previous studies in both mammalian (21, 45) and fish cells, where MAPK was activated by IGF-I-stimulated proliferation of embryonic cells (40). IGF-I tended to provoke higher increase of both MAPK-P and Akt-P levels than insulin: this is not contradictory with the TKA/binding ratio of receptors, and it is explained by the higher number of IGF-I receptors of the trout skeletal muscle cells in culture (8).
Total MAPK levels increased with differentiation of trout skeletal muscle cells, which may be linked to the reported increase in IGF-IR when these cells differentiate from myoblasts to myotubes (8); a greater number of receptors in differentiated cells could give, as a result, increased basal levels of these proteins, and therefore, the myogenic differentiation would yield a more IGF-I-responsive phenotype. This is in agreement with Al-Khalili et al. (1), who also suggested that increased total MAPK levels under basal conditions during human skeletal muscle cell differentiation could be related to a more insulin-responsive state. The lower response of Akt-P and especially of MAPK-P to peptides in the differentiated myotubes at day 11 could be a consequence of the higher levels of control total Akt and total MAPK proteins, and as such, the activation of phosphorylation of both proteins by peptides could be lower.
Concerning the PI3-kinase-Akt pathway, we found no alterations in total Akt levels in any of the studied stages when cells were incubated with insulin or IGF-I. However, as with MAPK, we did observe an increase in Akt levels throughout skeletal muscle cell development in culture. This increase in total Akt levels may be linked to the reported increase in IGF-I-stimulated glucose uptake levels when these cells differentiate to myotubes in culture (7). Calera and Pilch (5) also reported a marked increase in endogenous Akt protein levels during differentiation of Sol8 skeletal muscle cells, and Gonzalez et al. (16) described the increase of total Akt levels during differentiation of cultured mouse muscle cells. In addition, Altomare et al. (2) observed increased Akt mRNA levels during C2C12 muscle cell differentiation, and, more recently, Al-Khalili et al. (1) described increased total Akt levels under basal conditions during human skeletal muscle cell differentiation. Insulin and IGF-I stimulated Akt phosphorylation, and such activation was observed throughout all developmental stages, from myoblasts to myotubes, with IGF-I tending to be more active than insulin, particularly in myoblasts, which suggests that the PI3-kinase-Akt pathway is activated during all stages of trout skeletal muscle development. These results differ from data reported in other studies describing the PI3-kinase-Akt pathway as more active in differentiated myotubes (9, 42). Nevertheless, it has been reported that trout skeletal muscle cells in a primary culture can proliferate and differentiate at the same time, indeed from the earliest stages of in vitro development (14, 41), a fact that may explain why insulin and IGF-I activate the Akt signaling pathway in trout myoblasts. After these initial steps, trout cells continue to differentiate, with little proliferation. This may explain the lack of significant change in MAPK-P levels detected in peptide-stimulated fully differentiated myotubes (day 11), although changes in Akt-P levels were significant.
In conclusion, the present study describes not only the differing activity of purified insulin and IGF-I receptors in trout, but also the IGF-I- and insulin-signaling pathways throughout the in vitro development of rainbow trout skeletal muscle cells. The insulin receptor seems to be more active than IGF-IR in trout, as it exhibits higher tyrosine kinase activity per unit of receptor than IGF-IR. Moreover, IGF-I and insulin activate different intracellular signal transduction pathways in trout skeletal muscle cells, depending on the stage of cell culture development; in differentiating and proliferating myoblasts, both the MAPK and Akt pathways are activated by IGF-I and insulin, whereas in fully differentiated myotubes the peptides only activate the Akt pathway, although to a lower extent than in myoblasts.
| GRANTS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
Present address of Ina Ammendrup-Johnsen: Department of Pharmacology, The Panum Institute 18.6, Blegdamsvej 3, DK-2200, University of Copenhagen, Copenhagen N, Denmark.
| FOOTNOTES |
|---|
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 |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
I. Seiliez, J.-C. Gabillard, S. Skiba-Cassy, D. Garcia-Serrana, J. Gutierrez, S. Kaushik, S. Panserat, and S. Tesseraud An in vivo and in vitro assessment of TOR signaling cascade in rainbow trout (Oncorhynchus mykiss) Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2008; 295(1): R329 - R335. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Seiliez, S. Panserat, S. Skiba-Cassy, A. Fricot, C. Vachot, S. Kaushik, and S. Tesseraud Feeding Status Regulates the Polyubiquitination Step of the Ubiquitin-Proteasome-Dependent Proteolysis in Rainbow Trout (Oncorhynchus mykiss) Muscle J. Nutr., March 1, 2008; 138(3): 487 - 491. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |