| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH COMMUNICATION |
a Deutsches Krebsforschungszentrum, Division of Immunochemistry, D-69120 Heidelberg, Germany
b INSERM, U294 et Labo Immuno-Hemato, 75018 Paris, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: redox regulation • signal transduction • glutathione • CHO-HIR • BCNU
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Oxidizing substances such as thiol reactive agents, millimolar concentrations of hydrogen peroxide, or vanadate and pervanadate can exert insulin-like effects on intact cells (6–15). Whenever tested, these procedures involved the insulin-independent tyrosine phosphorylation of the IR ß-chain in vivo (11–14) and therefore did not clarify the events that precede phosphorylation of the tyrosine residues in the DFG loop.
Sulfhydryl-modifying reagents were found to either stimulate or inhibit intrinsic tyrosine kinase activity of the IR (15–18). An increase of basal IR kinase activity toward an exogenous peptide substrate was seen after treatment of permeabilized IR-transfected Chinese hamster ovary cells (CHO/HIR) cells with the thiol-reactive reagent maleimidobutyrylbiocytin (MBB) (16); the sulfhydryl alkylating agent iodoacetamide increased the catalytic activity of the cytoplasmic domain of the IR, whereas another alkylating agent, N-ethyl-maleimide (NEM), inhibited the activity (17, 18). These studies indicated strongly that the functional activity of the IR ß-chain can be modulated by oxidative modification or by derivatization of one or more of the cysteine residues of the IR ß-chain. However, they failed to show whether and how such a modification may play a role in the cellular response to insulin.
To address this question, we investigated whether the insulin responsiveness of the IR may be inhibited in intact cells by antioxidants and synergistically enhanced by a shift to (mildly) oxidizing intracellular conditions, and if so, whether oxidizing intracellular conditions in the absence of insulin may be associated with structural and functional changes in the absence of detectable tyrosine phosphorylation of the IR ß-chain. This question was among others of clinical interest, because a markedly increased insulin responsiveness has been found in HIV-infected patients (19) and an abnormally low glutathione level and altered redox state have been found in skeletal muscle tissue of SIV-infected macaques (20).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cells and cell culture conditions
CHO cells and CHO-HIR (22, 23) were cultured routinely in F12 medium (Gibco/BRL) containing 10% fetal calf serum, penicillin, and streptomycin, and were split twice weekly. Twenty-four hours before starting the experiment, the confluent cell layers were cultured in modified NCTC medium (24) under serum-free conditions. Insulin, the glutathione biosynthesis inhibitor BSO (25), and the glutathione reductase inhibitor BCNU (26, 27) were added to the cells as indicated.
Cell lysis and immunoprecipitation
At harvesting, the cells were washed with ice-cold phosphate-buffered saline (PBS) containing 0.4 mM EDTA and immediately lysed in ice-cold IR lysis buffer [50 mM Tris/HCl, pH 7.2, 150 mM NaCl, 1% Nonidet P40, 0.5% Na+-deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 mM EGTA, 25 mM NaF, 1 mM Na+-orthovanadate, 1 mM PMSF, 5 µg/ml leupeptin, and 2 µg/ml aprotinin]. After scraping the cells from the dishes, lysates were centrifuged at 12,000 g for 15 min in the cold and the supernatants were precleared with protein A/G agarose for 1 h at 4°C. The cleared supernatants were assayed for protein content according to the method of Lowry et al. (28). Aliquots with an equal amount of protein were incubated with either monoclonal mouse or polyclonal rabbit human IR ß-chain antibody overnight at 4°C, as indicated. Immunocomplexes were collected by incubating the samples with protein A/G agarose for 1 h in the cold, washed three times with IR lysis buffer, twice with IR phosphorylation buffer (20 mM HEPES, pH 7.2, 100 mM NaCl, 5 mM MgCl2, 5 mM MnCl2) containing 100 µM Na+ orthovanadate, and assayed for tyrosine kinase activity as described below.
In vitro phosphorylation (kinase assay)
Immunoprecipitates were washed and incubated in 15 µl IR phosphorylation buffer containing either 5 nM or 25 µM ATP (10 µCi 32P-
-ATP, NEN-DuPont), as described, and incubated for 20 min at 30°C. The reaction was stopped by boiling in Laemmli sample buffer and the proteins of the samples were separated by reducing SDS-polyacrylamide gel electrophoresis (PAGE) (7% gels) in glycine/Tris buffer, pH 8.3. The phosphorylated IR was detected by autoradiography or subjected to Western blotting and probed with specific antibody.
Western blotting and detection of tyrosine phosphorylation
After SDS-PAGE, the proteins were routinely blotted on nitrocellulose or transferred to PVDF membranes for phosphoamino acid analysis as described previously (29). To demonstrate tyrosine phosphorylation, the membranes were incubated for 1 h at room temperature in PBS containing 5% skim milk powder and subsequently incubated for 2 h with 0.2 µg/ml monoclonal phosphotyrosine antibody 4G10. The membranes were then washed and incubated with a goat anti-mouse antibody coupled to horseradish peroxidase. Tyrosine-phosphorylated proteins were detected with the ECL reagent kit (Amersham, Arlington Heights, Ill.). In some cases (see text), membranes were subsequently incubated in stripping buffer (62.5 mM Tris/HCl, pH 6.7, 2% SDS, 100 mM ß-mercaptoethanol) for 30 min at 50°C and then probed with polyclonal anti-IR ß-chain antibody plus goat-anti rabbit antibody coupled to alkaline phosphatase and NBT/BCIP as substrate for detection of IR.
Phosphoamino acid analysis
Phosphoamino acid analysis of in vitro phosphorylated IR ß-chain from insulin or BSO/BCNU-treated CHO-T cells was performed by 1-dimensional analysis (29, 30). Briefly, immunocomplexed and phosphorylated IR ß-chain was subjected to SDS-PAGE and transferred to a PVDF membrane. Radiolabeled IR ß-chain was located by autoradiography and excised. The protein was then hydrolyzed in 5.7 N HCl for 1 h at 110°C and subjected to electrophoresis on a thin-layer cellulose plate (Merck, Darmstadt, Germany). Phosphoserine, phosphothreonine, and phosphotyrosine (3 µg) were used as standards on a separate lane and localized with ninhydrin. The phosphoamino acid composition was identified by autoradiography.
Tryptic phosphopeptide analysis
Immunoprecipitated IR was subjected to in vitro phosphorylation. After SDS-PAGE and blotting, the 32P-labeled ß-subunit was identified by autoradiography and excised from the nitrocellulose membrane. The protein was digested with 10 µg of modified trypsin (Boehringer-Mannheim) in NaHCO3 (pH 8.2) at 37°C for 24 h. An additional 10 µg of modified trypsin was added and digestion was continued for an additional 12–20 h. Trypsinization was terminated by addition of sample buffer and boiling for 3 min.
The phosphopeptides were separated by Tricine/SDS-PAGE, as described (31), using a 20 cm acrylamide gel consisting of a 3% stacking, 10% spacing, and 16% resolving gel. After electrophoresis, the gels were sealed with plastic wrap and subjected to autoradiography at -80°C. The molecular weight of the phosphopeptides was estimated using Rainbow molecular weight markers (Amersham).
Analysis of the free sulfhydryl groups of the IR ß-chain
Untreated or BSO/BCNU-treated CHO-T cells were washed with ice-cold PBS containing 0.4 mM EDTA and immediately lysed in 1 ml IR lysis buffer. After incubation on ice for 2 min, 200 µM MBB was added and the lysates were incubated on ice for another 10 min (16). DTT (2 mM) was subsequently added to remove unbound MBB. The lysates were immunoprecipitated with polyclonal IR ß-chain antibody overnight; the immunocomplexes were washed three times with IR lysis buffer, twice with IR washing buffer (50 mM Tris/HCl, pH 7.5, 150 mM, 0.1% Nonidet P-40, 25 mM NaF, 1 mM Na+-orthovanadate), boiled in Laemmli sample buffer, and subjected to reducing SDS-PAGE, using 7.5% gels. The proteins were transferred to nitrocellulose and the membrane was incubated for 30–60 min in TBS-Tween-20 (0.05%) (TBS-T) containing 1% PVP. Finally, the membranes were incubated in 1% PVP/TBS-T with a 1:40000 dilution of streptavidin coupled to peroxidase for 1 h at room temperature; the changes in free protein sulfhydryl groups were detected by ECL, using the Amersham kit.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fig. 1).
|
Treatment of CHO-HIR cells with BSO and BCNU induces mildly oxidizing intracellular conditions and decreases the proportion of free sulfhydryl groups of the IR ß-chain
To investigate the effects of mildly oxidizing intracellular conditions on the redox state and functional activity of the IR ß-chain, we modulated the intracellular glutathione level and redox state of CHO-HIR cells with buthionine sulfoximine (BSO), a specific glutathione synthetase inhibitor (25), and/or BCNU, a known inhibitor of glutathione reductase (27). Both drugs increase the intracellular GSSG/GSH ratio (data not shown). BCNU was also shown to increase protein-S-thiolation through inhibition of dethiolation (32). To determine the effect of BSO/BCNU treatment of intact cells on the free sulfhydryl groups of the IR protein, we studied the reactivity of the IR ß-chain with the thiol-alkylating compound MBB (see ref 16). The experiments revealed a substantial MBB signal corresponding to the size of the IR ß-chain (see arrow in Fig. 2,
upper panel) in samples both from untreated (lane 2) and BSO/BCNU-treated cells (lane 3). However, the signal from BSO/BCNU-treated cells was markedly weaker than that from untreated cells, although probing of the same blot with an IR ß-chain-specific antibody (anti-ßHIR) revealed equal amounts of protein. No MBB signal was seen if the lysates were incubated with NEM before MBB (not shown), confirming the specificity of MBB for sulfhydryl groups. Moreover, no band was detectable when the IR-specific antibody was omitted (lane 1). Cumulative results from three different experiments are shown in Fig. 2, lower panel. A strong decrease in MBB signal was also observed after treatment with 500 µM BCNU for 30 min (not shown).
|
BSO/BCNU treatment of CHO/HIR cells alters the functional activity of the IR kinase in the absence of detectable tyrosine phosphorylation of the IR ß-chain
Treatment of CHO-HIR cells with BSO ( Fig. 3,
lane 3) or BSO plus BCNU (lane 4) did not induce any detectable tyrosine phosphorylation of the IR ß-chain, whereas insulin induced strong phosphotyrosine signals corresponding to the IR ß-chain and a 150 kDa protein that may be IRS-1 (lanes 2 and 5). However, BSO/BCNU treatment enhanced synergistically the tyrosine phosphorylation of the IR ß-chain in response to insulin (lane 5) and induced tyrosine phosphorylation of three unidentified proteins of 60, 65, and 120 kDa (lanes 4 and 5).
|
To characterize the functional properties of the IR ß-chain from the BSO/BCNU-treated cells in more detail, we analyzed its kinase activity in vitro. The experiments ( Fig. 4)
revealed that the IR ß-chain from BSO-treated (lane 3) or BSO plus BCNU-treated cells (lane 4) was strongly tyrosine phosphorylated if incubated with 25 µM ATP but not with 5 nM ATP. With 25 µM ATP, the degree of tyrosine phosphorylation was as strong as the phosphorylation of the IR ß-chain from insulin-treated cells (lanes 2 and 5). In view of the phosphotyrosine signals in 
Fig. 3, it is believed that the strong phosphotyrosine signals from the insulin-treated cells (lanes 2 and 5 in the upper left panel) result at least partly from the insulin-induced tyrosine phosphorylation prior to cell lysis. The analysis of 32P incorporation in the presence of 5 nM ATP ( Fig. 4, lower left panel) showed that the in vitro autophosphorylation of the IR ß-chain from BSO and BCNU-treated cells (lanes 3 and 4) was stronger than that of untreated control cells (lane 1) and weaker than that of insulin-treated cells (lanes 2 and 5). At 25 µM ATP, in contrast, the IR ß-chain from BSO- and BCNU-treated cells showed even stronger 32P incorporation than that of insulin-treated cells. BSO/BCNU treatment potentiated the insulin-induced 32P incorporation (lane 5). Phosphoamino acid analysis of the 32P-labeled IR ß-chain from BSO/BCNU-treated cells revealed a strong signal corresponding to phosphotyrosine, but no phosphoserine or phosphothreonine, a pattern similar to that of insulin-treated cells ( Fig. 5).
|
|
IR kinase activities from BSO/BCNU-treated and insulin-treated cells show distinct phosphorylation patterns
Six major autophosphorylation sites corresponding to different cytoplasmic regions of the IR ß-chain have been described previously (reviewed in ref 31). Tyrosine 972 is located close to the transmembrane region, tyrosine residues 1158, 1162, and 1163 are located in the regulatory domain, and tyrosines 1328 and 1334 are located near the COOH-terminal end (4). All these tyrosine residues become phosphorylated after insulin stimulation and have been mapped by tryptic digestion and 1-dimensional Tricine/SDS-PAGE (31). This method was applied to the IR from BSO/BCNU-treated cells to identify the sites that are phosphorylated in vitro. After in vitro phosphorylation, the IR was subjected to tryptic digestion and the labeled peptides were compared with the known peptide pattern from insulin-treated cells (31). We made sure that phosphorylation of the IR from the BSO/BCNU-treated cells ( Fig. 6,
upper panel, lanes c and d) was adequately increased compared with baseline 32P incorporation (lane a) or insulin-stimulated phosphorylation (lanes b and d), respectively. When the excised IR ß-chain bands were trypsinized and subjected to 1-dimensional Tricine-PAGE, we observed striking differences between the IR phosphopeptide patterns from BSO/BCNU-treated cells and insulin-stimulated cells. BSO/BCNU-treated cells yielded only two phosphopeptides corresponding to the juxtamembrane and the regulatory domains of the ß-chain ( Fig. 6; lower panel, lane c), whereas insulin stimulated cells yielded eight phosphopeptides (lane b) corresponding to the three major domains of the IR ß-chain, as indicated in the figure (see ref 31). No labeled phosphopeptide corresponding to the COOH-terminal or regulatory domain could be obtained from the IR ß-chain from untreated cells (not shown).
|
IR kinase activity from BSO/BCNU-treated cells shows a rapid decay in vitro: failure to demonstrate the participation of a phosphatase
The IR kinase activity from BSO/BCNU-treated cells decayed in vitro almost completely during a 30 min incubation ( Fig. 7;
compare lanes 3 and 5). To determine whether this decay may be mediated by a tyrosine-specific phosphatase, we added vanadate, a potent inhibitor of tyrosine phosphatases. Our experiments showed that the decay was not prevented by vanadate ( Fig. 7, lanes 4, 6, and 8).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The absence of significant tyrosine phosphorylation in the insulin receptor ß-chain from BSO/BCNU-treated cells suggests that the `primed state' may be an intermediate stage between the fully inactive state described in ref 4 and the fully activated and tyrosine-phosphorylated state described in ref 5. In functional terms, this state is intermediate because it is associated with a kinase activity clearly superior to the unprimed stage and superior even to the kinase activity of insulin-treated cells if tested in vitro with 25 µM ATP, but inferior to the insulin-induced kinase activity if tested at 5 nM ATP ( Fig. 4). Moreover, the phosphorylation pattern seen in tryptic peptide maps is clearly more restricted after BSO/BCNU treatment if compared with insulin treatment. Most important, BSO/BCNU treatment is insufficient by itself to induce a substantial degree of tyrosine autophosphorylation in the intact cell ( Fig. 3). With respect to the signaling of typically insulin-dependent functions, this redox-primed state cannot be more than an intermediate stage because important functions such as the stabilization of the catalytic activity (5), induction of insulin receptor internalization (33), and interaction with the downstream signal component IRS-1 (5) clearly require tyrosine phosphorylation of the ß-chain.
The structural basis for the process of redox priming is unknown. However, the functional changes observed, together with the decrease of free sulfhydryl groups ( Fig. 2) and absence of detectable tyrosine phosphorylation of the IR ß-chain ( Fig. 3), suggest that an oxidative modification of one of the cysteine residues of the ß-chain may cause a conformational change in the DFG loop that allows ATP to enter the adenyl binding site at least to some extent. This interpretation remains to be tested by analysis of the corresponding crystal structure.
The modulation of insulin receptor function by redox processes and thiol-reactive agents has already been the subject of numerous reports (6–18). However, these studies described the oxidative activation by hydrogen peroxide and other agents essentially as an alternative to the insulin-mediated activation and failed to describe a redox-mediated intermediate state as a prerequisite of optimal insulin responsiveness. Whenever the degree of tyrosine phosphorylation has been studied in this context, it was found that the oxidatively activated insulin receptor was already tyrosine phosphorylated to a degree similar to that observed in insulin-stimulated cells (11–14). The failure to detect a functionally altered but not yet tyrosine-phosphorylated intermediate stage in these earlier studies may be explained partly by the much more aggressive oxidative agents (vanadate, pervanadate, and millimolar concentrations of hydrogen peroxide), which may act not only on the intracellular domains of the insulin receptor but also on the extracellular domains, as suggested in ref 15. Another explanation might be that the oxidative activation of the IR kinase activity in the previous studies might have been studied under culture conditions that still contained small amounts of insulin (i.e., small amounts of serum), and therefore may have proceeded to tyrosine phosphorylation and full functional activation of the IR ß-chain.
The small sulfhydryl-modifying reagent iodoacetamide was found to stimulate the tyrosine kinase activities of the insulin receptor and several other tyrosine kinase species in vitro (16–18, 34, 35). This lends strong support to the assumption that the oxidative modification of one of the free sulfhydryl groups of the IR ß-chain may cause a structural and functional alteration and that similar mechanisms may apply to other signal receptors with tyrosine kinase activities. In some cases, the authors also tested the effect of iodoacetamide after, but not before, insulin treatment (16–18); therefore, they did not observe the priming effect in their studies. It is puzzling that the six different mutant proteins in which one of the six cysteine residues (i.e., Cys981, Cys1056, Cys1138, Cys1234, Cys1245, Cys1308) was mutated to alanine showed no loss of catalytic function (18). The structural basis for the observed change of functional activity thus requires further investigations.
The sensitivity of the insulin-induced kinase activation against antioxidants and the enhancement of insulin responsiveness by moderately oxidizing intracellular conditions suggest a physiologically important role of redox regulation in IR function and may explain the puzzling paradox that insulin binding at the extracellular 
-chains was found to induce the autophosphorylation of the ß-chain (1, 2), even though structural studies showed that the adenine binding site of the unactivated kinase domain is blocked (4). Our finding that the response to insulin is synergistically enhanced by BSO/BCNU treatment and inhibited by BHA and N-acetyl-cysteine suggests that the initial structural and functional changes in the IR ß-chain may be induced by a redox process. These redox-mediated changes may not be sufficient to trigger the downstream signal cascade, but may render the IR responsive to insulin.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: IR, insulin receptor; MBB, maleimidobutyrylbiocytin; NEM, N-ethyl-maleimide; BHA, butylated hydroxyanisole; BCNU, 1-(2-chloroethyl)-3-(2-hydroxyethyl)-1-nitrosourea; BSO, buthionine sulfoximine; PVDF, polyvinylidene difluoride; NBT, nitroblue tetrazolium; BCIP, 5-bromo-4-chloro-3-indolyl phosphate; CHO-HIR, Chinese hamster ovary cells stably transfected with the human insulin receptor; PMSF, phenylmethylsulfonyl fluoride; TBS-T, TBS-Tween-20 (0.05%); PBS, phosphate-buffered saline.
Received for publication December 22, 1997. Accepted for publication January 30, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-glutamylcysteine synthetase activity and abnormal sulfate levels in liver and muscle tissue may explain cysteine and glutathione levels in SIV-infected rhesus macaques. AIDS Res. Hum. Retrovir. 12, 1639–1641[Medline]
This article has been cited by other articles:
|
|
 |
|
  H. Z. Ozsoy, N. Sivasubramanian, E. D. Wieder, S. Pedersen, and D. L. Mann Oxidative Stress Promotes Ligand-independent and Enhanced Ligand-dependent Tumor Necrosis Factor Receptor Signaling J. Biol. Chem., August 22, 2008; 283(34): 23419 - 23428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Cuervo Calorie Restriction and Aging: The Ultimate "Cleansing Diet" J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2008; 63(6): 547 - 549. [Full Text] [PDF]
|
|
|
|
|
|
W. Droge Oxidative Aging and Insulin Receptor Signaling J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2005; 60(11): 1378 - 1385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Schmitt, A. Hotz-Wagenblatt, H. Klein, and W. Droge Interdependent Regulation of Insulin Receptor Kinase Activity by ADP and Hydrogen Peroxide J. Biol. Chem., February 4, 2005; 280(5): 3795 - 3801. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-K. Kwon, H.-J. Jang, S. Kole, H.-J. He, and M. Bernier Role of the pleckstrin homology domain of PLC{gamma}1 in its interaction with the insulin receptor J. Cell Biol., October 27, 2003; 163(2): 375 - 384. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Droge Free Radicals in the Physiological Control of Cell Function Physiol Rev, January 1, 2002; 82(1): 47 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Elbaz, K. Bedecs, M. Masson, M. Sutren, A. D. Strosberg, and C. Nahmias Functional Trans-inactivation of Insulin Receptor Kinase by Growth-Inhibitory Angiotensin II AT2 Receptor Mol. Endocrinol., June 1, 2000; 14(6): 795 - 804. [Abstract] [Full Text]
|
|
|
|
 |
|
 E Maellaro, S Dominici, B Del Bello, M. Valentini, L Pieri, P Perego, R Supino, F Zunino, E Lorenzini, A Paolicchi, et al. Membrane gamma-glutamyl transpeptidase activity of melanoma cells: effects on cellular H(2)O(2) production, cell surface protein thiol oxidation and NF-kappa B activation status J. Cell Sci., January 8, 2000; 113(15): 2671 - 2678. [Abstract] [PDF]
|
|
|
|
 |
|
 E. SCHMID, A. HOTZ-WAGENBLATT, V. HACK, and W. DRÖGE Phosphorylation of the insulin receptor kinase by phosphocreatine in combination with hydrogen peroxide: the structural basis of redox priming FASEB J, September 1, 1999; 13(12): 1491 - 1500. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Ushmorov, V. Hack, and W. Droge Differential Reconstitution of Mitochondrial Respiratory Chain Activity and Plasma Redox State by Cysteine and Ornithine in a Model of Cancer Cachexia Cancer Res., July 1, 1999; 59(14): 3527 - 3534. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
