Modulation of insulin-like growth factor actions in L6A1 myoblasts by insulin-like growth factor binding protein (IGFBP)-4 and IGFBP-5: A dual role for IGFBP-5

Download Modulation of insulin-like growth factor actions in L6A1 myoblasts by insulin-like growth factor binding protein (IGFBP)-4 and IGFBP-5: A dual role for IGFBP-5

Post on 06-Jun-2016

214 views

Category:

Documents

1 download

Embed Size (px)

TRANSCRIPT

  • JOURNAL OF CELLULAR PHYSIOLOGY 177:4757 (1998)

    Modulation of Insulin-Like Growth FactorActions in L6A1 Myoblasts by Insulin-LikeGrowth Factor Binding Protein (IGFBP)-4

    and IGFBP-5: A Dual Role for IGFBP-5DAINA Z. EWTON,1* SHARON A. COOLICAN,1 SUBBURAMAN MOHAN,2

    STEVEN D. CHERNAUSEK,3 AND JAMES R. FLORINI11Department of Biology, Syracuse University, Syracuse, New York

    2Department of Medicine, Biochemistry, and Physiology, Loma Linda University,Loma Linda, California

    3Department of Endocrinology, Childrens Hospital Medical Center, Cincinnati, Ohio

    We have previously shown that the insulin-like growth factors (IGFs) stimulateboth proliferation and differentiation of skeletal muscle cells in culture, and thatthese actions in L6A1 muscle cells may be modulated by three secreted IGFbinding proteins (IGFBPs), IGFBP-4, -5, and -6. Since we found that the temporalexpression pattern of IGFBP-4 and IGFBP-5 differed dramatically during the transi-tion from proliferating myoblasts to differentiated myotubes, we undertook thecurrent study to examine the effects of purified IGFBP-4 and IGFBP-5 on IGF-stimulated actions in L6A1 muscle cells. As has been shown for other cell types,we found that IGFBP-4 had only inhibitory actions, inhibiting IGF-I and IGF-II-stimulated proliferation and differentiation. In contrast, IGFBP-5 exhibited bothinhibitory and stimulatory actions. When added in the presence of 30 ng/ml IGF-I, IGFBP-5 (250 ng/ml) inhibited all markers of the early proliferative response:the tyrosine phosphorylation of the cytoplasmic signaling molecules IRS-1 andShc, the activation of the MAP kinases, ERK1 and 2, the elevation of c-fos mRNA,the early inhibition of the elevation in myogenin mRNA, and the increase in cellnumber. In contrast, IGFBP-5 stimulated all aspects of the myogenic response toIGF-I: the later rise in myogenin mRNA, the elevation of creatine kinase activity,and the fusion of myoblasts into myotubes. This dual response to IGFBP-5 wasgreatest when it was added at a molar ratio of IGFBP-5 to IGF-I of 2:1. In contrast,when IGFBP-5 was added in the presence of IGF-II, it inhibited both proliferationand differentiation. Neither IGFBP had any effect when added in the presenceof R3 IGF-I, an analog with substantially reduced affinity for IGFBPs. Our resultssuggest that the role of IGFBP-4 is mainly to sequester excess IGFs, and thusinhibit all actions. IGFBP-5, however, is capable of eliciting a dual response,possibly due to its unique ability to associate with the cell membrane. J. Cell.Physiol. 177:4757, 1998. q 1998 Wiley-Liss, Inc.

    The insulin-like growth factors (IGF-I and IGF-II) ily of IGF binding proteins (IGFBPs; reviewed inare unique among mitogens in that they not only stimu- Rajaram et al., 1997; Jones and Clemmons, 1995;late proliferation of skeletal muscle cells in culture, but Bach and Rechler, 1995). L6 muscle cells have beenthey also stimulate myogenic differentiation (reviewed shown to secrete three IGFBPs, IGFBP-4, -5, and -6in Florini et al., 1996). Even in the absence of exogenous (McCusker et al., 1989; McCusker and Clemmons,IGFs, they have an established role as autocrine/para- 1994; Ewton et al., 1994a; Silverman et al., 1995; Ew-crine modulators of skeletal muscle cell growth and ton and Florini, 1995). In our studies, we found thatdifferentiation. IGFs have been shown to be expressed each of these IGFBPs displayed a unique pattern ofby several muscle cell lines in culture (Tollefsen et al.,1989a,b; Florini et al., 1991a; Brown et al., 1992; Rosenet al., 1993; Magri et al., 1994). Evidence from studies

    Contract grant sponsor: NIH; Contract grant number: HL11551;measuring relative potencies of IGF-I and IGF-II and Contract grant sponsor: USDA; Contract grant number: 9603268.from studies with IGF analogs with altered receptor*Correspondence to: Dr. Daina Z. Ewton, Biology Department,affinities suggests that the type I IGF-I receptor medi-Syracuse University, 130 College Place, Syracuse, NY 13244.ates both proliferation and differentiation in L6A1E-mail: dzewton@mailbox.syr.edumuscle cells (Ewton et al., 1987, 1994b).

    The actions of IGFs are further modulated by a fam- Received 26 August 1997; Accepted 13 March 1998

    q 1998 WILEY-LISS, INC.

    JCP-10193/ 893f$$0193 07-20-98 20:16:31 wlcpa W Liss: JCP

  • EWTON ET AL.48

    expression during the various stages of L6A1 muscle sponse (Coolican et al., 1997). Our studies showed thatIGF-I treatment of L6A1 myoblasts resulted in thecell growth and differentiation (Ewton and Florini,

    1995). IGFBP-4 mRNA levels were highest in prolifer- early phosphorylation of both MAP kinase isoforms,ERK1 and ERK2. This phosphorylation was inhibitedating myoblasts but decreased during IGF-stimulated

    differentiation. No IGFBP-5 mRNA could be detected by the MEK inhibitor PD098059, resulting not only ininhibition of proliferation, but also in a dramatic in-in proliferating myoblasts, but levels increased dramat-

    ically during differentiation. IGFBP-6 mRNA levels in- crease in the subsequent myogenic response.The recent availability of purified IGFBPs promptedcreased in serum-free medium as cultures became qui-

    escent in the absence of IGFs. These observations sug- our study to investigate the actions of each of theIGFBPs secreted by L6A1 cells on the mitogenic andgested to us that each of the three IGFBPs may play

    a specialized role during myogenesis. In addition, ana- myogenic actions of the IGFs, including effects on theIGF-stimulated signal transduction pathway. Bach etlogs of IGF-I with reduced affinity for IGFBPs were

    about 10 times more potent than native IGF-I in stimu- al. (1994, 1995) have shown that purified IGFBP-6 in-hibited IGF-II-stimulated, but not IGF-I-stimulated,lating L6A1 proliferation, but about 100 times more

    potent in stimulating differentiation, suggesting that proliferation and differentiation of L6A1 myoblasts. Inthe present study, we describe the effects of purifiedthe IGFBPs secreted by the cells were inhibitory to

    these actions of IGF-I (Ewton et al., 1994a; Silverman IGFBP-4 and -5 on IGF-I and IGF-II-stimulated pro-cesses in L6A1 cells, and find that while IGFBP-4 iset al., 1995; Ewton and Florini, 1995).

    It is not obvious how the IGFs can stimulate two inhibitory under all conditions tested, IGFBP-5 ap-pears to have a dual role when added in the presencemutually exclusive processes such as proliferation and

    differentiation, or what role the secreted IGFBPs play of IGF-I, inhibiting all aspects of the proliferative re-sponse, but stimulating the myogenic response.in modulating the effects of the IGFs. It is also unclear

    how one receptor can mediate signaling for two suchMATERIALS AND METHODSantagonistic processes, and the question still remains:

    MaterialsWhere do the signaling pathways for IGF-stimulatedproliferation and differentiation diverge? It now seems Recombinant human IGF-I was a gift from Ciba-

    Geigy (Summit, NJ), and the analog R3 IGF-I was aclear that the stimulatory effects on proliferation anddifferentiation are temporally separated. Initially, IGF- gift from Paul Walton and John Ballard (GroPep Pty

    Ltd., Adelaide, Australia). Recombinant human IGF-I stimulates the proliferative response and inhibits themyogenic response; this is followed by withdrawal from II was kindly provided by Eli Lilly (Indianapolis, IN).

    IGFBP-5 was purified from human bone extract as pre-the cell cycle and upregulation of myogenic signals anddifferentiation (Ewton et al., 1994b; Rosenthal and viously described (Bautista et al., 1991; Mohan et al.,

    1995), and the homogeneity of the preparation was es-Cheng, 1995; Engert et al., 1996). In other cell types,the mitogenic signal transduction pathway from the tablished by sodium dodecyl sulfate-polyacrylamide gel

    electrophoresis (SDS-PAGE) and amino-terminal se-IGF-I receptor, a tyrosine kinase, has been shown toinvolve tyrosine phosphorylation of several intracellu- quence analysis. No detectable levels of IGF-I or IGF-

    II were found in the purified preparations of IGFBP-5.lar substrates (reviewed in Jones and Clemmons, 1995;LeRoith et al., 1995), including IRS-1 and Shc, the acti- Rat IGFBP-4 was isolated from media conditioned by

    the B104s cell line (Cheung et al., 1991) as previouslyvation of the Ras-Raf-MAP kinase pathway, followedby the activation of nuclear transcription factors (see described (Cheung et al., 1994). It contained mainly

    the 24 kDa species, but a small amount (10%) of 28review by LeRoith et al., 1995). In addition, phosphory-lated IRS-1 associates with PI 3-kinase, leading to the kDa glycosylated IGFBP-4 was present. Purity was ver-

    ified by silver staining after SDS-PAGE and by electro-downstream activation of p70 S6 kinase. Similar earlyevents associated with the mitogenic response to IGF- spray mass spectrometry which indicated a single spe-

    cies of correct mass. Tissue culture medium compo-I in muscle cells have also been described: stimulationof tyrosine phosphorylation of IRS-1 and Shc (Lamph- nents were purchased from Gibco/BRL (Grand Island,

    NY). The c-fos cDNA probe was obtained from ATCCere and Lienhard, 1992; Giorgino and Smith, 1995),induction of the nuclear transcription factor c-fos (Ong (Rockville, MD) and the myogenin cDNA probe was a

    gift from Eric N. Olson (University of Texas, South-et al., 1987; Giorgino and Smith, 1995), suppression ofmyogenin gene expression, inhibition of retinoblastoma western Medical Center, Dallas, TX). The anti-rat car-

    boxy-terminal IRS-1 polyclonal antibody, anti-humanprotein (Rb) dephosphorylation, and upregulation ofthe gene expression of cyclin-dependent kinase 4 and Shc polyclonal antibody, and recombinant Protein A

    agarose were from Upstate Biotechnology, Inc. (Lakecyclin D1 (Rosenthal and Cheng, 1995; Engert et al.,1996). This early mitogenic response is followed by Placid, NY). Horseradish peroxidase conjugated anti-

    phosphotyrosine monoclonal antibody RC20H was ob-stimulation of differentiation characterized by eleva-tion in the steady-state levels of myogenin mRNA (Flo- tained from Transduction Laboratories (Lexington,

    KY). The antibody to phospho-MAPK was purchasedrini et al., 1991b), an increase in the levels of creatinekinase (CK) activity, and fusion of myoblasts into myo- from New England Biolabs, Inc. (Beverly, MA). Goat

    anti-rabbit horseradish peroxidase conjugated IgG, thetubes (Ewton and Florini, 1981).By using specific inhibitors of signaling molecules of BCA protein assay reagents, and the SuperSignalTM

    CL-HRP substrate system for enhanced chemilumines-the IGF-I signal transduction pathway, we have re-cently shown that the activation of the MAP kinase cence were purchased from Pierce (Rockford, IL). The

    MEK inhibitor PD098059 was purchased from New En-pathway is necessary for IGF-I-stimulated L6A1 my-oblast proliferation, while the PI 3-kinase/p70 S6 ki- gland BioLabs. All other chemicals were from Sigma

    (St. Louis, MO).nase pathway plays a critical role in the myogenic re-

    JCP-10193/ 893f$$0193 07-20-98 20:16:31 wlcpa W Liss: JCP

  • ROLE OF IGFBPs IN MYOGENESIS 49

    dase conjugated anti-phosphotyrosine monoclonal anti-Cell culturebody RC20H (1:2500) in blocking buffer. The blots wereL6A1 myoblasts were grown in Dulbeccos modified washed six times for 5 minutes each in TBST at roomEagles medium (DMEM) containing 10% horse serum, temperature, and immunolabeling detected by en-1% chick embryo extract, and 1% antibiotic antimycotic hanced chemiluminescence according to the manufac-solution (Sigma). Cultures were plated at 4 1 104 cells/ turers directions. The blots were stripped by incuba-2 cm2 well for cell number and CK determination, and tion for 1 hour at 507C in 62.5 mM Tris-HCl, pH 6.7,at 45 1 105 cells/60 mm dish for RNA extraction or 2% SDS, and 0.7% -mercaptoethanol, washed for 1

    immunoprecipitation of proteins. After incubation hour at room temperature in four changes of TBST,overnight, the cultures were washed with DMEM be- and reblocked in blocking buffer containing 3% nonfatfore the addition of IGFs and/or IGFBPs in DMEM dry milk in TBST. The blots were immunolabeled withcontaining 0.1% bovine serum albumin (BSA). polyclonal antibodies to IRS-1 or Shc (1 mg/ml in TBST

    containing 3% nonfat dry milk) overnight at 47C, fol-Quantitation of proliferationlowed by washing and detection as described above.and differentiation

    For phospho-MAPK detection, total cell lysates wereTo quantitate cell proliferation, cells were trypsin- prepared by washing the cell monolayer twice with ice-

    ized and counted in a Model ZBI Coulter Counter 24 cold PBS followed by lysis in boiling 11 Laemmli sam-hours after the addition of growth factors. Differentia- ple buffer containing 100 mM DTT, 1 mM Na3VO4, 1tion was quantified 72 hours after the addition of mM NaF, and analyzed by SDS-PAGE and transferredgrowth factors by measuring the elevation in CK activ- to nitrocellulose as described above. The blots wereity using the NAD-coupled microtiter assay (Florini, blocked for 1 hour at room temperature in blocking1989). Statistical analysis was performed by using Stu- buffer containing 5% nonfat dry milk in PBST and incu-dents t-test. bated overnight at 47C with the antibody against phos-

    pho-MAPK (1:1,000) in 5% milk, PBST. After washingNorthern blot analyses six times in PBST, the membranes were incubated for 1hour at room temperature with horseradish peroxidaseTotal RNA was isolated from cultures using theconjugated goat-anti rabbit IgG diluted 1:3,000 in 5%RNeasy total RNA kit (Qiagen, Chatsworth, CA) ac-milk, PBST. The blots were washed six times in PBSTcording to the manufacturers instructions. Five to tenat room temperature, and immunolabeling was de-microgram samples were analyzed on Northern blots.tected as described above.Consistency of RNA loading was verified by visualiza-

    tion of ethidium bromide-stained ribosomal RNARESULTSbands. The 32P-labeled probe for c-fos mRNA was pre-

    Effect of IGFBP-4 on IGF-I-stimulated L6A1pared by random priming of the 1 kb Pst I restrictionmyoblast proliferation and differentiationfragment, and for myogenin, the 1.1 kb BamH I/EcoR

    I restriction fragment. Since IGFBP-4 is the major IGFBP secreted by prolif-erating L6 myoblasts (McCusker et al., 1989; McCuskerImmunoprecipitation and Western blottingand Clemmons, 1994; Ewton and Florini, 1995), we de-The cell monolayers in 60 mm dishes were washed termined its role in modulating the mitogenic and myo-twice with ice-cold phosphate-buffered saline (PBS) genic actions of IGF-I on L6A1 myoblasts. Figure 1Aand the cells were lysed by incubating the cultures for shows that the addition of exogenous IGFBP-4 to L6A130 minutes in 400 ml cold modified radioimmunoprecip- myoblasts in the presence of 30 ng/ml (4 nM) IGF-Iitation assay (RIPA) buffer (50 mM Tris-HCl, pH 7.4, gave a concentration-dependent inhibition of cell prolif-1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, eration. IGF-I alone caused almost a doubling (93%1 mM EGTA, 1 mM PMSF, 1 mg/ml aprotinin, 1 mg/...

Recommended

View more >