help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pandey, S. K.
Right arrow Articles by Bernier, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pandey, S. K.
Right arrow Articles by Bernier, M.
Endocrinology Vol. 143, No. 2 375-385
Copyright © 2002 by The Endocrine Society

INTRACELLULAR SIGNAL SYSTEMS

Wortmannin-Sensitive Pathway Is Required for Insulin-Stimulated Phosphorylation of Inhibitor {kappa}B{alpha}

Sanjay K. Pandey, Hua-Jun He, Alan Chesley, Magdalena Juhaszova, Michael T. Crow and Michel Bernier

Diabetes Section, Laboratory of Clinical Investigation (S.K.P., H.-J.H., M.B.), Laboratory of Cardiovascular Science (A.C., M.T.C.), and Research Resource Branch (M.J.), National Institute on Aging, NIH, Baltimore, Maryland 21224-6825

Address all correspondence and requests for reprints to: Michel Bernier, Ph.D., Diabetes Section, Laboratory of Clinical Investigation, National Institute on Aging, NIH, 5600 Nathan Shock Drive, Baltimore, Maryland 21224. E-mail: bernierm{at}vax.grc.nia.nih.gov


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The aim of this study was to examine the signaling pathways by which insulin promotes activation of nuclear factor {kappa}B (NF{kappa}B) through the regulation of inhibitor {kappa}B{alpha} (I{kappa}B{alpha}). We show here that although insulin increased {kappa}B-dependent reporter gene expression and augmented nuclear translocation of the p65/RelA subunit of NF{kappa}B and its DNA binding, it was able to induce a time-dependent accumulation of phosphorylated and ubiquitinated I{kappa}B{alpha} without its proteolytic degradation. In contrast, cell stimulation with the cytokine TNF{alpha} allowed activation of NF{kappa}B through phosphorylation, ubiquitination, and subsequent degradation of I{kappa}B{alpha}. Immunofluorescence studies revealed the presence of a large pool of phosphorylated I{kappa}B{alpha} in the nucleus of unstimulated and insulin-treated cells. I{kappa}B kinase {alpha} and ß, central players in the phosphorylation of I{kappa}B{alpha}, were rapidly induced following exposure to TNF{alpha} but not insulin. Furthermore, insulin-stimulated I{kappa}B{alpha} phosphorylation did not depend on activation of the Ras/ERK cascade. Expression of a dominant-negative mutant of Akt1 or class I PI3K inhibited the insulin stimulation of PI3K/Akt1 signaling without affecting phosphorylation of I{kappa}B{alpha}. Interestingly, the PI3K inhibitors wortmannin and LY294002 blocked insulin-stimulated class I PI3K-dependent events at much lower doses than that required to inhibit phosphorylation of I{kappa}B{alpha}. These data demonstrate that insulin regulates I{kappa}B{alpha} function through a distinct low-affinity wortmannin-sensitive pathway.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE NUMBER OF cells in tissues is determined by the homeostatic balance between proliferation of new cells and death of senescent or damaged cells. Apoptosis, also termed programmed cell death, is an active, genetically controlled process that has been identified as a key phenomenon is the pathogenesis of a wide variety of diseases, including diabetes (1). Although the biochemical and genetic events leading to apoptotic death have been the subject of intense investigation, some studies reported that insulin, through its binding to the insulin receptor (IR), confers protection against apoptosis mediated by the removal of serum and growth factors from the culture medium in a number of cell types, which include IR-expressing Chinese hamster ovary (CHO) cells (2, 3, 4, 5).

A recent report established that the survival function of insulin can occur by means of activation of nuclear factor {kappa}B (NF{kappa}B) through Raf-1 activation (6). The transcription factor NF{kappa}B, a p50/p65 heterodimer, is maintained in an inactive form in the cytoplasm by association with the inhibitory protein inhibitor {kappa}B{alpha} (I{kappa}B{alpha}) (7). I{kappa}B{alpha} possesses both a nuclear localization signal and a nuclear export sequence that allows it to shuttle between the cytoplasm and the nucleus (8). As recently shown, the nuclear export property of I{kappa}B{alpha} contributes to the largely cytoplasmic localization of inactive NF{kappa}B/I{kappa}B{alpha} complexes, thus allowing efficient NF{kappa}B activation by extracellular signals. On cell stimulation with the cytokine TNF{alpha} as well as a broad range of stimuli, I{kappa}B{alpha} is rapidly phosphorylated on two serine residues (Ser-32 and Ser-36) by the two main I{kappa}B kinase (IKK) activities, which then marks I{kappa}B{alpha} for ubiquitin-mediated proteolysis by the 26S proteasome complex (9). As a result of I{kappa}B{alpha} degradation, the released NF{kappa}B heterodimer is then free to translocate to the nucleus, in which it activates transcription of target genes including I{kappa}B{alpha}. There is evidence, however, for the existence of alternative NF{kappa}B activation pathways. Stimulus-induced tyrosine phosphorylation of I{kappa}B{alpha} allows its association with PI3K, thus causing dissociation of the inhibitory protein from NF{kappa}B (10), and intranuclear proteolysis of incoming I{kappa}B{alpha} avoids termination of nuclear NF{kappa}B activity during cell activation (11).

Remarkably, little is known about the mode of regulation of NF{kappa}B in response to insulin. Therefore, we undertook a study to examine the role of insulin on phosphorylation and degradation of I{kappa}B{alpha}, the primary regulator of signal-induced activation of NF{kappa}B, and to compare this effect with that of the cytokine TNF{alpha}, a prototypical NF{kappa}B activator. During the course of the study, it was observed that insulin stimulated phosphorylation of I{kappa}B{alpha} without promoting its proteolytic degradation in a way that TNF{alpha} did. It was therefore decided to extend this observation through an analysis of the insulin-signaling pathways that may be contributing to the regulation of I{kappa}B{alpha} function. In this study, a number of methods were employed, including the use of pharmacological inhibitors, indirect immunofluorescence, expression of active and inactive proteins, and Western blots using various phospho-specific antibodies.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
PD98059, LY294002, wortmannin, and insulin were purchased from Calbiochem (La Jolla, CA); U0129 from Promega Corp. (Madison, WI); cycloheximide from Sigma Chemical Co. (St. Louis, MO); and TNF{alpha} from R\|[amp ]\|D Systems (Minneapolis, MN). Protein G-Plus/protein A- agarose was obtained from Oncogene Science, Inc. (Manhasset, NY); electrophoresis reagents such as gels, Tris-glycine SDS running buffer, and polyvinylidene difluoride (PVDF) membrane were purchased from Novex Corp. (San Diego, CA); polyclonal antibodies to the phosphorylated forms of Akt (Ser-473) and I{kappa}B{alpha} (Ser-32) were purchased from New England Biolabs, Inc. (Beverly, MA). Glutathione S-transferase (GST)-I{kappa}B{alpha}, horseradish peroxidase-conjugated monoclonal antibody antiubiquitin antibody (clone P4D1), and rabbit polyclonal antibodies to c-Src and the phosphorylated form of PKC-{zeta} (Thr-410) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA); rabbit polyclonal antibodies to the phosphorylated form of GSK-3ß (Ser-9) was purchased from Cell Signaling Technology, Inc. (Beverly, MA). Anti-Flag M2 monoclonal antibody affinity gel was purchased from Kodak Co. (New Haven, CT) and Sigma; rabbit polyclonal antibody to c-myc was from Upstate Biotechnology, Inc. (Lake Placid, NY). The plasmids expressing myc epitope-tagged Akt1 were purchased from UBI, whereas the plasmid-encoding Flag-tagged I{kappa}B{alpha} was a kind gift from Dr. Albert S. Baldwin (University of North Carolina). The plasmid-encoding dominant- negative mutant form of the p85 subunit of class I PI3K({Delta}p85) was provided by Drs. W. Ogawa and Masato Kasuga (Kobe University, Japan). [{gamma}-32P]ATP (3000 Ci/mmol) was purchased from Amersham Pharmacia Biotech (Arlington Heights, IL).

Cell culture and transient transfection assays
CHO cells expressing the wild-type (WT) human IR (CHO-IR) and CHO-IR/{Delta}SOS cells that stably express a transdominant inhibitory mSOS1 mutant have been described elsewhere (12). Human HepG2 cells were purchased from ATCC (Manassas, VA). Cells were cultured in Ham’s F-12 medium (CHO cells) or DMEM with high glucose (HepG2 cells) supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin, and 10% FBS (Gemini, Calabasas, CA), and maintained in a humidified atmosphere of 5% CO2 in air at 37 C.

Transient cotransfection of CHO-IR cells was performed according to the manufacturer’s protocols for the use of TransFast reagent (Promega Corp.). In some experiments, basal and stimulated NF{kappa}B-mediated luciferase activity was analyzed according to the procedure of Bertrand et al. (6) with slight modification. Briefly, cells seeded into 35-mm dishes were transiently transfected in 800 µl Ham’s F-12 medium supplemented with 0.3% FBS, 0.1 µg pCMVSport ß-galactosidase (Life Technologies, Inc., Rockville, MD; as a monitor for transfection efficiency) and 0.5 µg pNF{kappa}B-TA-Luc (CLONTECH Laboratories, Inc., Palo Alto, CA), a reporter plasmid (pGL2) containing four tandem copies of the {kappa}B consensus sequence (5'-GGGAATTTCC-3') (13) inserted upstream of a minimal thymidine kinase promoter driving a luciferase reporter gene. One hour later, 1.5 ml Ham’s F-12 medium containing 0.3% serum was added to each dish. Twenty-four hours after transfection, cells were incubated in the absence or the presence of 10 nM insulin for 24 h or 20 ng/ml TNF{alpha} for 7 h before assaying luciferase activity, which was normalized on the basis of ß-galactosidase expression. Cell extract preparation as well as luciferase and ß-galactosidase assays were performed according to the manufacturer’s protocols (Promega Corp.). In some experiments, cells were transfected with promoterless pGL2 plasmid.

Preparation of nuclear extracts and EMSA
EMSA analysis of nuclear extracts prepared from untreated and insulin-treated cells was performed according to a recently described procedure (14). Briefly, CHO-IR cells were grown to confluency in 100-mm dishes and incubated in serum-free medium (SFM) for 3 h before treatment. Nuclear proteins were isolated 6 h after stimulation with 10 nM insulin by the method of Muller et al. (15), with minor modifications. In some experiments, 20 ng/ml TNF-{alpha} was added for 30 min. Gel-purified double-stranded consensus NF{kappa}B probe (5'-AGTTGAGGGGACTTTCCCAGG-3') (Promega Corp.) was end-labeled with [{gamma}-32P]ATP. Identical amounts of radioactive probe (1–2 x 105 cpm) were added to binding reactions containing 2 µg nuclear extracts in a final volume of 10 µl in DNA-binding buffer (Promega Corp.). Specificity was determined by prior addition of 100-fold excess unlabeled consensus NF{kappa}B or Oct-1 (TGTCGAATGCAAATCACTAGA) oligonucleotide. Where used, 2 µg of polyclonal rabbit NF{kappa}B p65 antibody (Santa Cruz Biotechnology) was incubated with nuclear extracts for 20 min at room temperature before the start of the reaction. Reaction mixtures were incubated for 20 min at 25 C before separating on nondenaturing 6% polycrylamide gels at room temperature and subjected to electrophoresis in 0.5x Tris-borate-EDTA buffer before vacuum drying and autoradiography. Protein determination was performed using the BCA protein assay reagent from Pierce Chemical Co. (Rockford, IL).

It is important to note that the rationale for using different times of insulin exposure was based on the premises that phosphorylation of I{kappa}B{alpha} precedes its ubiquitination, a step required for NF{kappa}B nuclear translocation. Accordingly, the EMSA analysis is routinely performed using nuclear extracts prepared from untreated cells and cells treated with insulin for 6 h (14). Similarly, a longer incubation period with insulin was used for the analysis of NF{kappa}B-dependent gene expression because of the time required for the transcriptional and translational processes to be activated in the cell.

Immunofluorescence microscopy
CHO-IR cells were seeded on glass coverslips and stained for indirect immunofluorescence. After fixing in fresh 4% paraformaldehyde in PBS supplemented with 200 nM okadaic acid (Calbiochem), cells were permeabilized in 0.1% (wt/vol) Triton X-100 in PBS for 10 min at room temperature, and then were incubated with blocking buffer (1% BSA in PBS) for 20 min. After a series of washes, cells were probed with 2 µg/ml of rabbit anti-phospho I{kappa}B{alpha} or 4 µg/ml of rabbit anti-I{kappa}B{alpha} for 16 h at 4 C. The primary antibodies were detected with AlexaFluor 488-conjugated antirabbit IgG (Molecular Probes, Eugene, OR). Cells were observed with an LSM-410 inverted confocal microscope (Carl Zeiss, Thornwood, NY). The confocal pinhole was set to obtain spatial resolution of 0.4 µm in the horizontal plane and 1 µm in the axial dimension.

In vitro IKK{alpha} and IKKß assay
The protocol for the in vitro kinase assay was adapted from the method of Uhlik et al. (16). After the indicated treatments, cells were lysed on ice in a lysis buffer containing 20 mM HEPES, pH 7.6, 250 mM NaCl, 1% Nonidet P-40 (wt/vol), 20 mM ß-glycerophosphate, 1 mM EDTA, 20 mM p-nitrophenylphosphate, 0.1 mM sodium orthovanadate, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and 1/100 volume of 100x protease inhibitor cocktail (Calbiochem). After 20 min on ice, lysates were clarified by centrifugation at 12,000 x g for 20 min at 4 C. The supernatants were used for immunoprecipitation of IKK{alpha} and IKKß with 2 µg/sample of rabbit polyclonal antihuman IKK{alpha} and IKKß (Santa Cruz Biotechnology) that were preadsorbed for 1 h with 100 µl of protein A/protein G-Sepharose slurry (50%). After a 3-h incubation at 4 C, the immunocomplexes were washed three times with lysis buffer, once with lysis buffer containing 1 M urea, and twice with the kinase reaction buffer (20 mM HEPES, pH 7.6, 20 mM MgCl2, 20 mM ß-glycerophosphate, 1 mM EDTA, 20 µM ATP, and 2 mM dithiothreitol). The kinase reaction was started by adding 30 µl of the kinase reaction buffer in the presence of 5 µCi of [{gamma}-32P]ATP and 1 µg of GST-I{kappa}B{alpha} (1–317) fusion protein per assay and incubating for 30 min at 30 C. The reactions were stopped by adding 13.5 µl of 3x Laemmli sample buffer.

In vitro c-Raf-1 kinase assay
The manufacturer’s protocols for the use of the c-Raf-1-coupled kinase assay kit was followed (Upstate Biotechnology, Inc.). Briefly, cells were lysed, and the clarified lysates were subjected to immunoprecipitation using 5 µg/assay of c-Raf-1 antibody preadsorbed to protein A/protein G-Sepharose. After a 2-h incubation at 4 C, the immunocomplexes were washed with lysis buffer and then incubated with 0.4 µg/assay of inactive ERK kinase (MEK)1 and 1 µg/assay of inactive ERK for 30 min at 30 C. The kinase reaction was started by adding 5 µCi of [{gamma}-32P]ATP and 10 µl of myelin basic protein (MBP) (2 mg/ml) per assay and incubating for 10 min at 30 C. The reactions were stopped by spotting samples onto p81 paper, which was then washed three times with 0.75% phosphoric acid and once with acetone before the determination of incorporated radioactivity by liquid scintillation counting.

In vitro Akt kinase assay
The manufacturer’s protocol for the use of the Akt kinase assay kit was followed (Cell Signaling Technologies). Briefly, CHO-IR cells were transiently transfected with an empty expression vector alone or with plasmids encoding WT or dominant negative mutant of myc-tagged Akt1. After a 3-h serum starvation, cells were left untreated or treated with 10 nM insulin for 2 h. Cells were lysed and the Akt kinase assay was performed on c-myc immunoprecipitates in the presence 200 µM ATP and 1 µg GSK-3 fusion protein for 30 min at 30 C. The kinase reaction was stopped by the addition of Laemmli sample buffer. The detection of phosphorylated GSK-3 was carried out by immunoblotting technique using phospho-specific antibodies.

Gel electrophoresis and Western blot analysis
Unless otherwise indicated, cells were lysed directly in Laemmli sample buffer containing 5% 2-mercaptoethanol and 1 mM vanadate. After heating at 70 C for 10 min, proteins were separated by SDS-PAGE on 4–12% polyacrylamide gradient gel along with prestained protein markers, and electrotransfer onto PVDF membrane. The membrane was incubated with blocking buffer (5% [wt/vol] nonfat dried milk in Tris-buffered saline 0.1% [wt/vol] Tween-20) for 1 h at room temperature and then probed with various antibodies for 1 h at room temperature or overnight at 4 C. After a series of washes, positive signals were visualized by chemiluminescence in combination with Hyperfilm-ECL (Amersham Pharmacia Biotech). The same membranes were then reprobed if needed. When justified, autoradiography of the membrane was performed to detect 32P-incorporation into substrates. Band intensities were quantitated by laser densitometry using the ImageQuant software (Molecular Dynamics, Inc., Sunnyvale, CA). The linearity of the signals on the blots was assessed as follows: Varying amounts of cell lysates were immunoblotted with a panel of antibodies and the ECL signal generated was then plotted. All experimental values felt within the linear portion of the curve.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Modulation of NF{kappa}B nuclear translocation, DNA binding, and {kappa}B-dependent transcription
A luciferase reporter gene assay was used to investigate whether the prosurvival function of insulin (2) was correlated with induction of {kappa}B-dependent transcription in CHO-IR cells. As shown in Fig. 1AGo, addition of insulin to SFM-treated cells for 24 h resulted in a 3.1 ± 0.5-fold increase in {kappa}B promoter activity (P < 0.001). A similar increase in {kappa}B-dependent transcription (3.6 ± 0.3-fold; P < 0.001) was observed on stimulation with TNF{alpha} (20 ng/ml). Control expression of the promoterless luciferase gene remained unchanged by the various treatments (data not shown).



View larger version (34K):
[in this window]
[in a new window]
  		Figure 1. Modulation of NF-{kappa}B activity in CHO-IR cells. A, Cells transiently cotransfected with 4x-{kappa}B-Luc reporter plasmid and pCMVSport ß-galactosidase expression plasmid were maintained for 24 h in Ham’s F-12 medium containing 0.3% FBS and then in SFM in the absence or the presence of insulin (10 nM) or TNF{alpha} (20 ng/ml) for an additional 24 h. Cell extracts were analyzed for luciferase activity and normalized for ß-gal as described under Materials and Methods. Results are the means ± SD of two experiments performed in duplicate dishes (n = 4). ***, P < 0.001 when compared with control. B, EMSA analysis in nuclear extracts from CHO-IR cells stimulated with insulin or TNF{alpha} for 6 h. C, Immunoblotting of the same nuclear extracts. Accumulation of p65 protein was detected using specific antibodies. Data are representative of three (panel B) or two (panel C) independent experiments.

 
To examine further the effect of these agents, DNA-binding EMSA and NF{kappa}B nuclear translocation analysis were performed. Consistent with the luciferase activity, a 6-h incubation with insulin induced a 5.3 ± 0.6-fold increase in NF{kappa}B DNA binding when compared with control unstimulated cells (Fig. 1BGo, left panel). The specificity of the complex was shown by competing out the signal with a 100-fold excess of unlabeled oligonucleotide containing the binding site for NF{kappa}B but not that of the unrelated Oct-1 sequence (data not shown). Incubation of nuclear extracts from insulin-treated cells with anti-p65 antibody markedly reduced the intensity of the radioactive band (Fig. 1BGo, right panel). The latter finding suggests the presence of p65 in this DNA-protein complex. Immunoblot analysis of nuclear proteins from CHO-IR cells treated in the presence of insulin or TNF{alpha} revealed an accumulation of the p65 protein in this fraction (Fig. 1CGo).

Having demonstrated the ability of insulin to stimulate NF{kappa}B, the subsequent studies were focused on the effects of insulin toward the regulation of I{kappa}B{alpha} phosphorylation and stability.

Insulin-induced I{kappa}B{alpha} phosphorylation without subsequent degradation
To study the effects of insulin on the phosphorylation of I{kappa}B{alpha}, we measured the accumulation of I{kappa}B{alpha} phosphorylated at Ser-32 by Western blot analysis of total cell lysates with a phospho-specific antibody. Treatment of CHO-IR cells with 10 nM insulin resulted in a time-dependent increase in I{kappa}B{alpha} phosphorylation that reached a maximum after 2 h stimulation and subsequent decline thereafter (Fig. 2AGo). The gradual decrease in phospho-I{kappa}B{alpha} signals observed following the 6-h time point is unlikely the result of a proteolytic degradation of endogenous pools of I{kappa}B{alpha} protein. A small but reproducible increase in the phosphorylation of I{kappa}B{alpha} was observed following cell stimulation for 10 min (Fig. 2BGo, right panel). The inhibition of protein synthesis with cycloheximide did not block phosphorylation and stability of I{kappa}B{alpha} in response to a 1-h or 3-h stimulation with insulin (Fig. 2CGo). Next, we evaluated whether TNF{alpha} would display similar kinetics of phosphorylation in our system. In cells treated with TNF{alpha}, the induction of I{kappa}B{alpha} phosphorylation occurred at a much faster rate, with maximum effect seen after 10 min and subsequent decline in I{kappa}B{alpha} protein to near undetectable levels (Fig. 2BGo, left panel). Longer exposure to TNF{alpha} (2–3 h) restored the levels of I{kappa}B{alpha} protein but was completely abolished when cells were pretreated with cycloheximide (Fig. 2DGo). Similar studies were carried out in a second insulin-responsive cell type. Like CHO-IR cells, the phosphorylation rate of I{kappa}B{alpha} in response to insulin was slower than with TNF{alpha} in HepG2 cells (Fig. 2EGo). The levels of I{kappa}B{alpha} protein were reduced after a 30-min stimulation with TNF{alpha}, whereas the I{kappa}B{alpha} content was intact in cells treated with insulin for periods up to 2 h. Reprobing the membrane with anti-ERK1/2 antibodies indicated equal loading of proteins in each lane.



View larger version (52K):
[in this window]
[in a new window]
  		Figure 2. Signal-induced phosphorylation and expression of I{kappa}B{alpha}. A, SFM-treated CHO-IR cells were incubated in the presence of insulin (10 nM) for the times indicated. B, Shorter kinetics were performed in response to TNF{alpha} (10 ng/ml) or insulin (10 nM). C, SFM-treated cells were preincubated with cycloheximide (10 µg/ml) for 30 min and then stimulated with insulin for 1 and 3 h. D, Following a 30-min preincubation with cycloheximide, CHO-IR cells were treated with TNF{alpha} (20 ng/ml) for 0–3 h. E, HepG2 cells were left untreated or were treated with insulin (10 nM) or TNF{alpha} (10 ng/ml) for the indicated times. In all instances, cell lysates were resolved by SDS-PAGE and analyzed for I{kappa}B{alpha} phosphorylation by Western blotting using phospho-specific antibodies to this protein. The membrane was then reprobed with an antibody against the total I{kappa}B{alpha} protein. Either c-Src (C and D) or ERK1/2 (E) protein levels were determined by immunoblotting with specific antibodies, as a loading control. Data in each panel are representative of three to four independent experiments.

 
The presence of phosphorylated I{kappa}B{alpha} in the absence of proteolytic degradation in insulin-treated cells is distinct from the mechanism of action of TNF{alpha}. The accumulation of phospho-I{kappa}B{alpha} promoted by insulin could be explained by the inhibition of ubiquitination, which is a prerequisite for its subsequent degradation by the proteasome. Alternatively, insulin may block proteasome-dependent protein degradation. To allow further study of I{kappa}B{alpha}, CHO-IR cells were pretreated with the peptide proteasomal inhibitor (PSI) that causes the accumulation of ubiquitinated proteins (17) and blocks activation of NF{kappa}B (18).

Our control experiment showed that TNF{alpha} rapidly induced I{kappa}B{alpha} phosphorylation as assessed either with anti-(pSer32) I{kappa}B{alpha} antibody (Fig. 3AGo, middle panel, lane 2) or the detection of the slowly migrating form of I{kappa}B{alpha} characteristic of the Ser32-Ser36-phosphorylated species (Fig. 3AGo, bottom panel, lane 2). Concomitantly, there is accumulation of immunoreactive phosphorylated I{kappa}B{alpha} protein that migrated as high-molecular-weight species (Fig. 3AGo, top panel, lane 2). As expected, TNF{alpha} exerted a time-dependent destabilization of phosphorylated I{kappa}B{alpha} (Fig. 3AGo, bottom panel, lanes 3–5). However, pretreatment of the cells with PSI maintained the levels of phospho-I{kappa}B{alpha} protein despite stimulation with TNF{alpha} for periods up to 2 h (Fig. 3AGo, bottom panel, lanes 8–10). Moreover, a significant increase in the pool of high-molecular-weight I{kappa}B{alpha} species was observed in PSI-treated cells in response to TNF{alpha} (Fig. 3AGo, top panel, lanes 8–10). The presence of polyubiquitinated forms of I{kappa}B{alpha} was confirmed by immunoprecipitation (Fig. 3CGo). However, it should be noted that the stabilization and accumulation of phosphorylated I{kappa}B{alpha} on TNF{alpha} stimulation was paralleled by a reduction of the unmodified pool of the protein (Fig. 3AGo, lower panel, lanes 8–10).



View larger version (46K):
[in this window]
[in a new window]
  		Figure 3. PSI increases phosphorylation and stabilization of ubiquitinated I{kappa}B{alpha}. CHO-IR cells were pretreated in the absence (lanes 1–5) or the presence (lanes 6–10) of 50 µM PSI for 1 h followed by the addition of 10 ng/ml TNF{alpha} (A) or 10 nM insulin (B) for the indicated times (0–120 min). Total cell lysates were immunoblotted with anti-phospho-I{kappa}B{alpha} antibody (upper panels, long exposure; middle panels, short exposure). Comparable aliquots were probed with the anti-I{kappa}B{alpha}, followed by anti-c-Src to show comparable protein load in each lane. The I{kappa}B{alpha} protein levels were quantitated and expressed as fold stimulation over control cells (time = 0). All blots and quantitative analysis are representative of two independent experiments. An asterisk denotes high-molecular-weight phosphorylated I{kappa}B{alpha}. C, CHO-IR cells were transiently transfected with an expression plasmid encoding Flag-tagged I{kappa}B{alpha}, serum-starved, and left untreated (-) or treated (+) with 50 µM PSI for 1 h before the addition of TNF{alpha} (10 ng/ml) or insulin (10 nM) for the indicated times. Ubiquitination of I{kappa}B{alpha} was detected by immunoprecipitation with anti-Flag antibody, followed by immunoblotting with the anti-ubiquitin antibody. The data are representative of two independent experiments.

 
In the presence of PSI, the amount of phosphorylated and polyubiquitinated I{kappa}B{alpha} increased with insulin in a time-dependent fashion (Fig. 3BGo, lanes 6–10) at a rate that was slower than that induced by TNF{alpha}. We could not detect significant reduction of the pool of I{kappa}B{alpha} after insulin treatment (Fig. 3BGo, bottom panel). It should be noted that PSI markedly increased the extent of I{kappa}B{alpha} phosphorylation without its degradation (Fig. 3Go, A and B, lane 1 vs. 6); interestingly, very little accumulation of polyubiquitinated forms of I{kappa}B{alpha} was observed. This would argue that phosphorylation alone is not sufficient to cause ubiquitination of I{kappa}B{alpha}. Taken together, these experiments suggest that the phosphorylation and ubiquitination of I{kappa}B{alpha} occur normally in response to insulin.

Immunofluorescence studies
To examine whether insulin and TNF{alpha} affect the localization of I{kappa}B{alpha}, indirect immunofluorescence microscopy was performed. As shown in Fig. 4AGo, phosphorylated I{kappa}B{alpha} demonstrated prominent nuclear staining in the presence or absence of stimulation by insulin under conditions in which unmodified I{kappa}B{alpha} was predominantly detected in the cytoplasm. Interestingly, a 2.2- and 3-fold increase in phospho-I{kappa}B{alpha} staining was observed in the cytosol of PSI- and TNF{alpha}-treated cells, respectively (Fig. 4BGo). When the same cells were further analyzed, the nuclear content of phosphorylated I{kappa}B{alpha} went up about 1.8-fold following a 5-min treatment with TNF{alpha}.



View larger version (94K):
[in this window]
[in a new window]
  		Figure 4. Cellular distribution of I{kappa}B{alpha} by indirect immunofluorescence. A, CHO-IR cells were left untreated (a, c) or treated with 10 nM insulin (b, d) for 4 h. Staining with anti-phospho-I{kappa}B{alpha} (upper panels) and anti-I{kappa}B{alpha} (lower panels) was performed on fixed cells as described in Materials and Methods. Similar results were obtained with HepG2 cells (data not shown). B, CHO-IR cells were incubated in the absence or the presence of 50 µM PSI for 1 h or 20 ng/ml TNF{alpha} for 5 min. Staining with anti-phospho-I{kappa}B{alpha} was then performed. Bars, 15 µm. The cells shown are representative of more than 100 cells. The data in each panel are representative of three independent experiments.

 
Effect of insulin on the activation of IKK{alpha} and IKKß
Because the I{kappa}B{alpha} kinases (IKK{alpha} and ß) constitute the key regulatory enzymes in phosphorylating the I{kappa}B{alpha} protein both at Ser-32 and Ser-36, it is plausible that the insulin-induced increase in I{kappa}B{alpha} phosphorylation might occur via a rise in IKK activity. To our surprise, using an immunoprecipitation-based kinase assay with GST-I{kappa}B{alpha} (1–317) fusion protein as the substrate, it was found that treatment of CHO-IR cells with insulin (2 h) failed to induce IKK activities, compared with control unstimulated cells (Fig. 5Go). In contrast, exposure to TNF{alpha} resulted in IKK activation. Thus, it is likely that the insulin-stimulated phosphorylation at Ser-32 of I{kappa}B{alpha} may result from the activation of additional enzyme(s) present in the insulin signaling pathways.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 5. Differential effect of insulin and TNF{alpha} on IKK activation. CHO-IR cells were treated in the absence or the presence of insulin (10 nM) or TNF{alpha} (20 ng/ml) for 2 h. Following incubation, IKK{alpha} and IKKß immunoprecipitates were prepared and then resuspended in a kinase reaction buffer containing 20 µM [{gamma}-32P]ATP and 1 µg GST-I{kappa}B{alpha} for 30 min at 30 C. After quenching the phosphorylation reaction, radiolabeled samples were resolved by SDS-PAGE, transferred onto PVDF membranes, and analyzed by autoradiography. Data are representative of three experiments.

 
Role of the Ras-ERK pathway in I{kappa}B{alpha} phosphorylation
Because Raf-1 is a key regulatory enzyme of the Ras/Raf-1/MEK/ERK cascade and has been shown to be involved in NF{kappa}B activation on insulin treatment (14), we examined whether expression of a transdominant negative mutant of the guanine nucleotide exchange factor, SOS1 protein, would ultimately have an effect on insulin-dependent I{kappa}B{alpha} phosphorylation in CHO-IR cells. To address this question, we first evaluated the activation of Raf-1 in a standard immunoprecipitation-based kinase assay in which CHO-IR/{Delta}SOS cells were incubated in the absence or the presence of insulin for 5 min. As shown in Fig. 6AGo, insulin-stimulated c-Raf-1 activity was totally absent in these cells. However, despite the lack of Ras-Raf-1 activation, phosphorylation without degradation of I{kappa}B{alpha} occurred in response to insulin (Fig. 6BGo). Pharmacological inhibition of MEK-1 with PD98059 (25 µM) or U0126 (1 µM) did not alter insulin’s ability to induce I{kappa}B{alpha} phosphorylation (data not shown). Of importance, insulin (1–10 nM) suppresses apoptosis mediated by the removal of serum and growth factors in CHO-IR/{Delta}SOS cells (15). Therefore, it would appear that signaling through the Ras/Raf-1/MEK/ERK pathway has little if any role in mediating insulin-dependent regulation of I{kappa}B{alpha} and antiapoptotic function.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 6. Raf-1-independent modulation of I{kappa}B{alpha} phosphorylation by insulin. A, c-Raf-1 immunoprecipitates from control and insulin-treated CHO-IR cells and CHO-IR/{Delta}SOS cells were resuspended in kinase reaction buffer containing 5 µCi radiolabeled ATP, inactive MEK-1, inactive ERK1/ERK2, and 20 µg MBP. Bars represent densitometry of phosphorylated MBP. Data are representative of a single experiment that has been performed twice with similar results. B, SFM-treated CHO-IR/{Delta}SOS cells were stimulated with insulin (10 nM) for the indicated times. Cell lysates were analyzed for both phospho-I{kappa}B{alpha} and total I{kappa}B{alpha} by Western blot technique using specific antibodies. This blot is representative of three experiments.

 
Role of the PI3K/Akt pathway in I{kappa}B{alpha} phosphorylation
Considering the newly described role of Akt (PKB) in NF{kappa}B activation (19, 20), we used molecular and pharmacological approaches to evaluate the importance of Akt in insulin-dependent I{kappa}B{alpha} phosphorylation. To this purpose, transient expression of myc-tagged Akt1 constructs in CHO-IR cells was carried out. In the first set of experiments, marked phosphorylation of glycogen synthase kinase-3ß (GSK-3ß) was observed in vitro in anti-myc immunoprecipitates from insulin-treated cells expressing wild-type (WT)-Akt1 construct (Fig. 7AGo). As anticipated, the dominant negative (DN) Akt1 (Lys179->Met) was unable to promote insulin-stimulated phosphorylation of GSK-3 (Fig. 7AGo). The levels of expression of WT- and DN-Akt1 constructs were comparable. In a second set of experiments, the consequences for insulin-induced Flag-tagged I{kappa}B{alpha} phosphorylation of coexpressing myc-tagged Akt1 constructs was investigated. Treatment of cells with insulin resulted in the accumulation of phosphorylated Flag-tagged I{kappa}B{alpha}, even in cells transfected with DN-Akt1 mutant (Fig. 7BGo), which indicates that this response is likely because of an Akt1-independent event.



View larger version (41K):
[in this window]
[in a new window]
  		Figure 7. Analysis of Akt1 constructs in CHO-IR cells. Cells were transiently transfected with a control plasmid (pUSEamp+) or expression vectors for WT or DN mutant Akt1 construct. A, Transfected cells were left untreated or incubated with insulin (10 nM) for 2 h. c-Myc immunoprecipitates were resuspended in kinase reaction buffer containing 200 µM ATP and 1 µg GSK3-fusion protein as the substrate. The samples were resolved by SDS-PAGE and immunoblotted with antiphospho-GSK3 antibody (upper panel) and anti c-myc antibody (lower panel). B, CHO-IR cells were cotransfected with Flag-tagged I{kappa}B{alpha} and plasmid DNA expressing Akt1 constructs. Serum-starved cells were stimulated with insulin for 2 h, and the anti-Flag immunoprecipitates were immunoblotted with phospho-specific I{kappa}B{alpha} antibody (upper panel) and then with anti-I{kappa}B{alpha} (lower panel). The experiments shown in A and B were repeated independently three to four times.

 
Class I PI3K is known for its sensitivity to low concentrations of the unrelated pharmacological inhibitors, wortmannin and LY294002 (21). The addition of wortmannin to CHO-IR cells led to a dose-dependent inhibition of insulin-stimulated phosphorylation of Akt at Ser-473, an indicator of its activation (22), with an IC50 of about 14 nM (Fig. 8Go, A and B). Similarly, LY294002 strongly inhibited phosphorylation of Akt with an IC50 of about 3 µM (data not shown). On the basis of these results, we next examined the effect of two concentrations of wortmannin (50 nM and 1 µM) on insulin-stimulated phosphorylation events. Through activation of PI3K, insulin promotes the Ser/Thr phosphorylation of several downstream protein kinases, which include GSK-3ß on Ser-9 and the atypical PKC-{zeta} on Thr-410 (23). Immunoblotting total lysates with phospho-specific antibodies demonstrated that insulin was able to transiently phosphorylate GSK-3ß and PKC-{zeta} reaching a maximum after 30 min of stimulation and a subsequent decline by 2 h (Fig. 8CGo, lanes 3–5). Insulin-stimulated phosphorylation of GSK-3ß and PKC-{zeta} was strongly inhibited by 50 nM wortmannin, a pattern similar to that of Akt phosphorylation (Fig. 8CGo, lanes 6 and 7). In contrast, phosphorylation of I{kappa}B{alpha} by insulin showed much lower sensitivity to wortmannin inhibition (Fig. 8CGo, bottom right panel). Quantitation of these data demonstrated that the maximum insulin-stimulated phosphorylation of I{kappa}B{alpha} was 3.2-fold after a 2-h incubation. Pretreatment of the cells with 50 nM wortmannin maintained insulin’s ability to phosphorylate I{kappa}B{alpha} by 2.2-fold, whereas addition of 1 µM wortmannin caused a small, but not statistically significant, increase in insulin response when compared with the control (Fig. 8CGo, lane 5 vs. 7 and 9). In parallel, reprobing the membranes with an anti-c-Src antibody showed equal amounts of proteins loaded in each lane. Phosphorylation of I{kappa}B{alpha} by insulin was clearly reduced with 30 µM LY294002 but not with 3 µM of the inhibitor (data not shown). It should be noted that treatment of cells with TNF{alpha} failed to promote significant phosphorylation of PKC-{zeta}, Akt, and GSK-3ß despite marked increase in phosphorylation of I{kappa}B{alpha} (Fig. 8CGo, lanes 1 and 2).



View larger version (53K):
[in this window]
[in a new window]
  		Figure 8. Insulin-stimulated I{kappa}B{alpha} phosphorylation does not require Class I PI3K. CHO-IR cells were treated with the indicated concentrations of wortmannin for 30 min before stimulation with 10 nM insulin for 2 h. A, Lysates were analyzed by immunoblotting with antibodies against phosphorylated Akt at Ser-473. B, Following band quantitation by densitometry, change in the levels of phosphorylated Akt was calculated and is presented as relative phosphorylation signal of the insulin-stimulated cells without wortmannin (set equal to 1). C, CHO-IR cells were treated with vehicle or wortmannin (50 and 1000 nM) for 30 min before stimulation with insulin for 30 and 120 min. In another set of cells, TNF{alpha} (10 ng/ml) was added for 5 min. Immunoblotting of cell lysates was performed using phospho-specific antibodies against PKC-{zeta} (pThr-410), Akt (pSer-473), GSK-3ß (pSer-9), and I{kappa}B{alpha} (pSer-32). Comparable aliquots were probed with anti-c-Src to show comparable protein load in each lane. The levels of phosphorylated I{kappa}B{alpha} were quantitated and expressed as fold stimulation over control cells (time = 0). Comparable results were obtained in three (A, B) or two (C) independent experiments. D and E, CHO-IR cells were transiently transfected with empty vector DNA or with the indicated plasmid DNAs (2 µg of myc-tagged Akt, 2 µg Flag-tagged I{kappa}B{alpha} with and without 3 µg {Delta}p85-PI3K), serum-starved, and then stimulated with 10 nM insulin for 2 h. Phosphorylation of Akt was detected by immunoprecipitation with c-myc antibody, followed by immunoblotting with phospho-specific Akt (D, upper panel) and anti-c-myc (D, lower panel) antibodies. E, Phosphorylation of I{kappa}B{alpha} was detected in anti-Flag immunoprecipitates as indicated in the legend of Fig. 7BGo. The immunoblots shown in D and E are representative of three independent experiments.

 
To assess directly whether class I PI3K played a prominent role in mediating insulin-induced I{kappa}B{alpha} phosphorylation, a dominant negative construct of the regulatory subunit of PI3K, p85, ({Delta}p85) was transiently cotransfected with Flag-tagged I{kappa}B{alpha} and myc-tagged Akt1 plasmids in CHO-IR cells. The {Delta}p85 mutant lacks the ability to bind the catalytic 110-kDa subunit of class I PI3K, thus causing the disruption of a subset of insulin-stimulated responses (24). Immunocomplexes from untreated and insulin-treated cells were prepared with antibodies against c-myc and Flag and immunoblotted (Fig. 8Go, D and E). {Delta}p85 Expression clearly reduced insulin-mediated phosphorylation of Akt (Fig. 8DGo). By contrast, expression of {Delta}p85 did not reduce basal and insulin-stimulated I{kappa}B{alpha} phosphorylation (Fig. 8EGo). These data support the hypothesis that Class I PI3K is not required for the insulin-stimulated I{kappa}B{alpha} phosphorylation.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
NF{kappa}B activity is modulated largely through phosphorylation and proteolytic degradation of the inhibitory protein I{kappa}B{alpha} in response to a number of external stimuli, including TNF{alpha} (25). However, the existence of alternative mechanisms for NF{kappa}B activation has also been described (26, 27, 28). In the present study, we investigated the signaling pathways by which insulin promotes NF{kappa}B activity through the regulation of I{kappa}B{alpha}. Insulin appears to be involved in the control of NF{kappa}B in cells from diverse origins, such as rat hepatoma cells, mouse skeletal C2C12 muscle cell line, and the human vascular endothelial cells (29, 30, 31). Our present study is the first to show that insulin triggers NF{kappa}B signaling through IKK-independent mechanisms, which involves Ser-32 phosphorylation and ubiquitination of I{kappa}B{alpha} without its degradation. Importantly, the resistance of phosphorylated I{kappa}B{alpha} to proteolytic degradation was observed in two insulin-responsive cell models, CHO-IR and HepG2 cells. Based on these findings, we hypothesized that a nonclassical IKK complex may account for phosphorylation of I{kappa}B{alpha} by insulin. Indeed, a key observation of this study is that I{kappa}B{alpha} phosphorylation and induction of NF{kappa}B DNA binding activity can be increased by insulin without detectable activation of IKK{alpha} or IKKß. The lack of involvement of the IKK complex in this response of insulin contrasts sharply from our observation that cells treated with TNF{alpha} are able to activate NF{kappa}B signaling by IKK-dependent mechanisms, as has been shown in a number of cell types by others. There is already evidence that the IKK complex is not involved in NF{kappa}B nuclear translocation and transactivation in response to a transcription coactivator or UV radiation (32, 33, 34). This raises the question of which kinase(s) could account for the phosphorylation of I{kappa}B{alpha} by insulin.

Our immunofluorescence microscopy data showed that both the constitutive and inducible I{kappa}B{alpha} phosphorylation occurred preferentially in the nucleus of CHO-IR and HepG2 cells. Implicit in this model is the ability of I{kappa}B{alpha} protein to enter the nucleus via its intrinsic nuclear import sequence (35) and be efficiently phosphorylated by a resident nuclear kinase. In this regard, IKK{alpha} and IKKß can be found both in the cytoplasm and the nucleus under certain experimental conditions (36). Recently, a new I{kappa}B kinase complex has been cloned (37), thus opening the possibility that this or another new kinase could account for phosphorylation of I{kappa}B{alpha} in the cell nucleus, either constitutively or upon cell activation.

An interesting observation is that insulin increases the phosphorylation-dependent I{kappa}B{alpha} ubiquitination in the absence of proteolytic degradation. This result indirectly indicates that the modified I{kappa}B{alpha} was recognized by ß-TrCP protein, the receptor for the SCF E3 ubiquitin ligase (38, 39) but failed to be marked for proteolysis. A recent study shows that ubiquitin-dependent degradation by the 26S proteasome is strongly inhibited by insulin in HepG2 cells (40). Therefore, insulin-stimulated block of the multicatalytic activity of nuclear 26 S proteasome (41) may promote sustained level of phosphorylated I{kappa}B{alpha} in the cell nucleus. By contrast, the marked depletion in the nuclear content in I{kappa}B{alpha} in response to TNF{alpha} (He, H.-J., and M. Bernier, unpublished results) may be because of its inability to inhibit proteosomal activity.

The mechanism by which insulin stimulates NF{kappa}B despite lack of I{kappa}B{alpha} degradation is not known. Because I{kappa}B{alpha} ubiquitination per se does not permit dissociation of the inhibitor present in the NF{kappa}B·IkB{alpha} complex (25), two mechanisms capable of releasing NF{kappa}B must be considered: phosphorylation of I{kappa}B{alpha} at new sites and/or interaction of I{kappa}B{alpha} with signaling molecules. For example, activation of the Ras/Raf-1/MEK/ERK cascade stimulates the activity of the family of 90-kDa ribosomal S6 kinases whose cellular functions include phosphorylation of the N-terminal regulatory domain of I{kappa}B{alpha} (42). However, the data presented here show the ability of insulin to increase phosphorylation of I{kappa}B{alpha} on Ser-32 (without its degradation) in CHO-IR/{Delta}SOS cells despite impaired Ras signaling (12), consistent with the notion that this pathway is not likely to play a major role. In a number of cell lines, insulin stimulates casein kinase II (43) and the atypical PKC-{zeta} ( 44, 45), two kinases known to phosphorylate I{kappa}B{alpha} protein on sites located in the Ankyrin repeat 6 and C-terminal PEST domain, respectively (46). Moreover, a subclass of Ras proteins has been recently shown to regulate the degradation of I{kappa}B protein (47). Our data do not support the possibility that the atypical PKC-{zeta}, may function as the protein kinase responsible for phosphorylation of I{kappa}B{alpha} by insulin because of the ability of low concentrations of wortmannin and LY294002 to block insulin-stimulated phosphorylation of PKC-{zeta} on Thr-410 without inhibitory effect on I{kappa}B{alpha} phosphorylation. Although GSK-3ß can function as an inducer of NF{kappa}B-dependent expression of a subset of survival genes (48), the ability of insulin to decrease GSK-3ß activity through phosphorylation on Ser-9 (49, 50) is consistent with a model wherein this protein-serine kinase has no significant effect on insulin-stimulated I{kappa}B{alpha} phosphorylation. Others have reported that the tyrosine phosphatase inhibitor pervanadate can activate NF{kappa}B transcriptional activity through tyrosine phosphorylation of I{kappa}B{alpha} without its proteolytic degradation, thus coupling NF{kappa}B to cellular tyrosine kinase (10). A recent study has also established that tyrosine phosphorylation events provide a key determinant for the binding of PI3K on I{kappa}B{alpha} that may underlie the release of NF{kappa}B to the nucleus in response to pervanadate (26). Nevertheless, the physiological significance of these observations is not clear. Whether tyrosine phosphorylation of I{kappa}B{alpha} is induced as part of an insulin-stimulated response toward NF{kappa}B-activity is unknown.

Conflicting results have begun to emerge regarding the role of Akt as a regulator of IKK catalytic activity. The apparent discrepancy is certainly linked to the fact that the involvement of Akt in NF{kappa}B signaling is cell type and stimulus specific. Consistent with this idea, activation of NF{kappa}B by TNF{alpha} is mediated by a PI3K/Akt pathway in some cell types (19, 51) but not others (20, 52). Our data clearly demonstrate the inability of TNF{alpha} to induce Akt despite marked IKK activation, I{kappa}B{alpha} phosphorylation and degradation, and induction of NF{kappa}B binding activity in CHO-IR cells. Moreover, we have done experiments in HepG2 cells and found efficient regulation of I{kappa}B{alpha} by this cytokine without induction of Akt.

Another observation of this study is that pharmacological inhibition of Class I PI3K with low concentrations of wortmannin or LY294002 blocked insulin-stimulated Akt phosphorylation and activity but did not translate into reduction in phosphorylation of I{kappa}B{alpha} by insulin. Furthermore, the dominant negative {Delta}p85 PI3K mutant also prevented insulin-dependent phosphorylation of Akt but not that of I{kappa}B{alpha}. Our observation that a kinase-negative mutant of Akt1 failed to disrupt phosphorylation of I{kappa}B{alpha} in response to insulin stimulation suggests that endogenous Akt1 activity was not necessary for this reaction to proceed. Thus, it would appear that Class I PI3K and Akt1 are unlikely players in linking activated insulin receptors with the phosphorylation of I{kappa}B{alpha}. As alluded earlier, low nanomolar concentrations of wortmannin inhibit Class I PI3K activity, whereas high concentrations of this drug can block the function of different classes of PI3K, including Class II PI3K, DNA-dependent protein kinase, and ataxia telangiectasia mutation (53, 54). A recent study reported that ataxia telangiectasia mutation participates in insulin signaling (55); hence, this multifunctional protein kinase may trigger activation of IKK-like enzymes with subsequent phosphorylation of I{kappa}B{alpha} by insulin.

Together, our results show that insulin promotes I{kappa}B{alpha} phosphorylation on Ser-32 in an atypical way: the phosphorylation of I{kappa}B{alpha} is mediated through a nonclassical IKK complex and is accompanied by ubiquitination without proteolytic degradation. The mechanisms underlying these events are unclear. Can the enzymes of the ubiquitination pathway promote different fates for I{kappa}B{alpha} in response to insulin and TNF{alpha}? Several ubiquitin-like proteins have recently been discovered that form conjugates with a number of cellular proteins but do not target them to the proteasome for degradation (56). Possibly, these relatives of ubiquitin may regulate the activity of NF{kappa}B through selective modification of I{kappa}B{alpha} in response to insulin. This function of insulin is independent of Ras- and Akt1-stimulated pathways but appears to require a class of PI3K enzymes that has low sensitivity toward wortmannin.


   Acknowledgments
 
We thank Drs. Albert S. Baldwin, Jr., Masato Kasuga, and Motoyoshi Sakaue for the plasmids and a cell line used in this study and Sutapa Kole for excellent technical assistance.


   Footnotes
 
Abbreviations: CHO, Chinese hamster ovary; CHO-IR, CHO cells expressing the wild-type human IR; DN, dominant negative; GSK, glycogen synthase kinase; GST, glutathione S-transferase; I{kappa}B, inhibitor {kappa}B; IKK, I{kappa}B kinase; IR, insulin receptor; MBP, myelin basic protein; MEK, ERK kinase; NF{kappa}B, nuclear factor {kappa}B; PSI, proteasomal inhibitor; PVDF, polyvinylidene difluoride; SFM, serum-free medium; WT, wild-type.

Received August 1, 2001.

Accepted for publication October 12, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hetts SW 1998 To die or not to die: an overview of apoptosis and its role in disease. JAMA 279:300–307[Abstract/Free Full Text]
  2. Lee-Kwon W, Park D, Baskar PV, Kole S, Bernier M 1998 Antiapoptotic signaling by the insulin receptor in Chinese hamster ovary cells. Biochemistry 37:15747–15757[CrossRef][Medline]
  3. Dudek H, Datta SR, Franke TF, Birnbaum MJ, Yao R, Cooper GM, Segal RA, Kaplan DR, Greenberg ME 1997 Regulation of neuronal survival by the serine-threonine protein kinase Akt. Science 275:661–665[Abstract/Free Full Text]
  4. Yenush L, Zanella C, Uchida T, Bernal D, White MF 1998 The pleckstrin homology and phosphotyrosine binding domains of insulin receptor substrate 1 mediate inhibition of apoptosis by insulin. Mol Cell Biol 18:6784–6794[Abstract/Free Full Text]
  5. Diaz B, Serna J, De Pablo F, de la Rosa EJ 2000 In vivo regulation of cell death by embryonic (pro)insulin and the insulin receptor during early retinal neurogenesis. Development 127:1641–1649[Abstract]
  6. Bertrand F, Atfi A, Cadoret A, L’Allemain G, Robin H, Lascols O, Capeau J, Cherqui G 1998 A role for nuclear factor {kappa}B in the antiapoptotic function of insulin. J Biol Chem 273:2931–2938[Abstract/Free Full Text]
  7. Baldwin Jr AS 1996 The NF-{kappa}B and I{kappa}B proteins: new discoveries and insights. Annu Rev Immunol 14:649–683[CrossRef][Medline]
  8. Johnson C, Van Antwerp D, Hope TJ 1999 An N-terminal nuclear export signal is required for the nucleocytoplasmic shuttling of I{kappa}B{alpha}. EMBO J 18:6682–6693[CrossRef][Medline]
  9. May MJ, Ghosh S 1998 Signal transduction through NF-{kappa}B. Immunol Today 19:80–88[CrossRef][Medline]
  10. Imbert V, Rupec RA, Livolsi A, Pahl HL, Traenckner EB, Mueller-Dieckmann C, Farahifar D, Rossi B, Auberger P, Baeuerle PA, Peyron JF 1996 Tyrosine phosphorylation of I{kappa}B-{alpha} activates NF-{kappa}B without proteolytic degradation of I{kappa}B-{alpha}. Cell 86:787–798[CrossRef][Medline]
  11. Renard P, Percherancier Y, Kroll M, Thomas D, Virelizier JL, Arenzana-Seisdedos F, Bachelerie F 2000 Inducible NF-{kappa}B activation is permitted by simultaneous degradation of nuclear I{kappa}B{alpha}. J Biol Chem 275:15193–15215[Abstract/Free Full Text]
  12. Park D, Pandey S, Maksimova E, Kole S, Bernier M 2000 Akt-dependent antiapoptotic action of insulin is sensitive to farnesyltransferase inhibitor. Biochemistry 39:12513–12521[CrossRef][Medline]
  13. Pessara U, Koch N 1990 Tumor necrosis factor alpha regulates expression of the major histocompatibility complex class II-associated invariant chain by binding of an NF-{kappa}B-like factor to a promoter element. Mol Cell Biol 10:4146–4154[Abstract/Free Full Text]
  14. Bertrand F, Philippe C, Antoine PJ, Baud L, Groyer A, Capeau J, Cherqui G 1995 Insulin activates nuclear factor {kappa}B in mammalian cells through a Raf-1-mediated pathway. J Biol Chem 270:24435–24441[Abstract/Free Full Text]
  15. Muller MM, Schreiber E, Schaffner W, Matthias P 1989 Rapid test for in vivo stability and DNA binding of mutated octamer binding proteins with "mini-extract" prepared from transfected cells. Nucleic Acids Res 17:6420[Free Full Text]
  16. Uhlik M, Good L, Xiao G, Harhaj EW, Zandi E, Karin M, Sun SC 1998 NF-{kappa}B-inducing kinase and I{kappa}B kinase participate in human T-cell leukemia virus I Tax-mediated NF-{kappa}B activation. J Biol Chem 273:21132–21136[Abstract/Free Full Text]
  17. Figueiredo-Pereira ME, Berg KA, Wilk S 1994 A new inhibitor of the chymotrypsin-like activity of the multicatalytic proteinase complex (20S proteasome) induces accumulation of ubiquitin-protein conjugates in a neuronal cell. J Neurochem 63:1578–1581[Medline]
  18. Griscavage JM, Wilk S, Ignarro LJ 1996 Inhibitors of the proteasome pathway interfere with induction of nitric oxide synthase in macrophages by blocking activation of transcription factor NF-{kappa}B. Proc Natl Acad Sci USA 93:3308–3312[Abstract/Free Full Text]
  19. Ozes ON, Mayo LD, Gustin JA, Pfeffer SR, Pfeffer LM, Donner DB 1999 NF-{kappa}B activation by tumour necrosis factor requires the Akt serine-threonine kinase. Nature 401:82–85[CrossRef][Medline]
  20. Romashkova JA, Makarov SS 1999 NF-{kappa}B is a target of AKT in anti-apoptotic PDGF signalling. Nature 401:86–90[CrossRef][Medline]
  21. Vanhaesebroeck B, Alessi DR 2000 The PI3K-PDK1 connection: more than just a road to PKB. Biochem J 346:561–576
  22. Hemmings BA 1997 Akt signaling: linking membrane events to life and death decisions. Science 275:628–630[Free Full Text]
  23. Pessin JE, Saltiel AR 2000 Signaling pathways in insulin action: molecular targets of insulin resistance. J Clin Invest 106:165–169[Medline]
  24. Hara K, Yonezawa K, Sakaue H, Ando A, Kotani K, K