FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.07-8173com.
 (The FASEB Journal. 2007;21:3696-3704.)
© 2007 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.07-8173comv1
21/13/3696    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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 reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Öst, A.
Right arrow Articles by Strålfors, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Öst, A.
Right arrow Articles by Strålfors, P.

Retinol-binding protein-4 attenuates insulin-induced phosphorylation of IRS1 and ERK1/2 in primary human adipocytes

Anita Öst*,1, Anna Danielsson*,1, Martin Lidén{dagger}, Ulf Eriksson{dagger}, Fredrik H. Nystrom{ddagger} and Peter Strålfors*,2

* Department of Cell Biology and Diabetes Research Centre, Linköping University, Linköping, Sweden;

{dagger} Ludwig Institute for Cancer Research, Stockholm, Sweden; and

{ddagger} Department of Medicine and Care and Diabetes Research Centre, Linköping University, Linköping, Sweden

2Correspondence: Department of Cell Biology and Diabetes Research Centre, Linköping University, SE58185 Linköping, Sweden. E-mail: peter.stralfors{at}ibk.liu.se


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Reduced sensitivity to insulin in adipose, muscle, and liver tissues is a hallmark of type 2 diabetes. Animal models and patients with type 2 diabetes exhibit elevated levels of circulating retinol-binding protein (RBP4), and RBP4 can induce insulin resistance in mice. However, little is known about how RBP4 affects insulin signaling. We examined the mechanisms of action of RBP4 in primary human adipocytes. RBP4-treated adipocytes exhibited the same molecular defects in insulin signaling, via IRS1 to MAP kinase, as in adipocytes from patients with type 2 diabetes. Without affecting autophosphorylation of the insulin receptor, RBP4 blocked the insulin-stimulated phosphorylation of IRS1 at serine (307) [corresponding to serine (302) in the murine sequence] and concomitantly increased the EC50 (from 0.5 to 2 nM) for insulin stimulation of IRS1 phosphorylation at tyrosine. The phosphorylation of IRS1 at serine (312) [corresponding to serine (307) in the murine sequence] was not affected in cells from diabetic patients and was also not affected by RBP4. The EC50 for insulin stimulation of downstream phosphorylation of MAP kinase ERK1/2 was increased (from 0.2 to 0.8 nM) by RBP4. We show that ERK1/2 phosphorylation is similarly impaired in adipocytes from patients with type 2 diabetes. However, the sensitivity to insulin for downstream signaling to control of protein kinase B and glucose uptake was not affected by RBP4. When insulin-resistant adipocytes from patients with type 2 diabetes were incubated with antibodies against RBP4, insulin-induced phosphorylation of IRS1 at serine (307) was normalized and the EC50 for insulin stimulation of ERK1/2 phosphorylation was reduced. Endogenous levels of RBP4 were markedly reduced in adipocytes from obese or type 2 diabetic subjects, whereas expression levels of RBP4 mRNA were unaffected. These findings indicate that RBP4 may be released from diabetic adipocytes and act locally to inhibit phosphorylation of IRS1 at serine (307), a phosphorylation site that may integrate nutrient sensing with insulin signaling.—Öst, A., Danielsson, A., Lidén, M., Eriksson, U., Nystrom, F. N., Strålfors, P. Retinol-binding protein-4 attenuates insulin-induced phosphorylation of IRS1 and ERK1/2 in primary human adipocytes.


Key Words: insulin resistance • type 2 diabetes • adipokine • protein phosphorylation • MAP kinase


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
INSULIN CONTROLS TARGET CELLS BY BINDING to cell surface receptors, which autophosphorylate on tyrosine residues to provide binding sites for downstream docking proteins that in turn are phosphorylated on tyrosine residues (reviewed in ref. 1 ). The insulin receptor substrate (IRS) proteins have been shown to be such docking proteins of insulin signaling. In adipocytes, IRS1 is directly tyrosine phosphorylated by the insulin receptor and transmits the insulin signal further downstream via a number of signaling mediators, including protein kinase B/Akt, eventually regulating glucose uptake and other metabolic effects. In human fat cells, IRS has been found to also mediate mitogenic signaling of insulin, via MAP kinase control of the transcription factor Elk-1 (2) . Insulin resistance is due to a defect in the signal transduction, which can be overcome by increased concentrations of insulin and enhanced insulin receptor activation. The pancreatic ß-cells can thus compensate insulin resistance by releasing more insulin. Eventually ß-cell failure often occurs and type 2 diabetes can be diagnosed.

In primary human adipocytes from patients with type 2 diabetes, the first step to being insulin resistant in insulin signal transduction is the insulin receptor-catalyzed tyrosine phosphorylation of IRS1 (3) . Adipocytes from these patients require three times higher concentration of insulin to elicit half-maximal tyrosine phosphorylation of IRS1 than cells from nondiabetic subjects. This is linked to attenuation of insulin-induced phosphorylation of IRS1 on serine (307) [corresponding to serine (302) in the murine sequence] in cells from diabetic patients (4) . Phosphorylation of serine (307) by insulin corresponds to an increased steady-state level of tyrosine phosphorylation of IRS1 in response to physiological concentrations of insulin (4 , 5) . This is in contrast to phosphorylation of IRS1 at serine (312) [corresponding to serine (307) in the murine sequence], which acts to inhibit the tyrosine phosphorylation of IRS1 in response to insulin in cells lines and mice (reviewed in refs. 6 7 8 9 ) and seemingly also in human adipocytes (5) . The role of serine (312) phosphorylation in the diabetic state in humans has not been investigated though. Enhanced basal phosphorylation/activation of MAP kinase ERK1/2 in adipocytes from patients with type 2 diabetes has been reported (10) , while the effect of insulin on the maximal phosphorylation of ERK1/2 was reported not to be affected in skeletal muscle from type 2 diabetic patients (11) . The sensitivity to insulin for phosphorylation of ERK1/2 in fat cells from patients with type 2 diabetes has, however, not been reported.

Adipose tissue has been found to synthesize and release a range of hormonal substances—adipokines— of which several, e.g., adiponectin, tumor necrosis factor-{alpha}, and resistin, have been found to affect insulin sensitivity in different cell types. Adipose-specific knockout of insulin-regulated glucose transporter (GLUT4) results in animals that are insulin resistant also in the skeletal muscle tissue (12) . It has recently been found that these animals have increased retinol-binding protein (RBP4) mRNA in the adipose tissue, with elevated levels of serum RBP4. The protein is circulating at elevated concentrations in several animal models of type 2 diabetes and insulin resistance (13) . Injection or over-expression of RBP4 in mice also induces insulin resistance (13) , suggesting a causal relationship between RBP4 and insulin resistance. In vitro differentiated and cultured human adipocytes secrete RBP4 (14) , and in human serum RBP4 levels positively correlate with the degree of insulin resistance and are associated with components of the metabolic syndrome, while exercise training reduces RBP4 levels and enhances the sensitivity to insulin (15) . However, the positive correlation of RBP4 with insulin resistance has been challenged, as it was recently reported (14) that subcutaneous adipocytes from obese subjects exhibit lower RBP4 gene expression than cells from lean subjects.

Herein, we report on the mechanisms of action of RBP4 in human adipocytes. Overnight incubation of primary human adipocytes with RBP4 faithfully mimicked the situation in type 2 diabetes at the level of IRS1 phosphorylation at tyrosine and at serine (307), as well as the downstream phosphorylation of MAP kinases ERK1/2, while the phosphorylation of protein kinase B and sensitivity of glucose uptake stimulation was not affected by RBP4. We also report that in adipocytes from patients with type 2 diabetes insulin signaling to increased phosphorylation of IRS1 at serine (307) and to downstream MAP kinase ERK1/2 was restored after treatment of the diabetic cells with anti-RBP4 antibodies.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
Informed consent was obtained from all participating individuals, and the procedures were approved by the local ethics committee. Subcutaneous fat was obtained from elective abdominal surgery on patients (see figure legends) during general anesthesia. Subjects were recruited consecutively from elective surgery at the departments of Obstetrics and Gynecology, and Surgery, at the University Hospital in Linköping. The only selection criterion for nondiabetic subjects was that they had not been diagnosed with diabetes. Patients with diagnosed type 2 diabetes were selected if they conformed to criteria for the metabolic syndrome: obesity/overweight [body mass index (BMI)>28] and dyslipidemia and/or hypertension. We did not match control subjects for obesity or BMI (see figure legends).

Materials
Mouse antiphosphotyrosine (PY20) monoclonal antibodies were from Transduction Laboratories (Lexington, KY, USA). Rabbit phospho-threonine (308)-protein kinase B and anti-IRS1 polyclonal antibodies were from Upstate Biotech (Charlottesville, VA, USA). Antiphospho-serine (307)-IRS1, antiphospho-serine (312)-IRS1, and phospho-ERK1/2 polyclonal antibodies from Cell Signaling Techn. (Beverly, MA, USA). Rabbit polyclonal antiactin antibodies were from Santa Cruz Biotechn. (Santa Cruz, CA, USA). Mouse monoclonal anit-GLUT4 antibodies were from Biogenesis (Poole, UK). Purified L48 mouse anti-RBP4 monoclonal antibodies was obtained and characterized as described previously (16) . Human RBP4, insulin, and other chemicals were from Sigma-Aldrich (St. Louis, MO, USA) or as indicated in the text. About 50% of the RBP4 carried the ligand retinol as determined by the A325/A280 absorbance ratio (17) .

Isolation and incubation of adipocytes
Adipocytes were isolated from subcutaneous adipose tissue by collagenase (type 1, Worthington, NJ, USA) digestion as described previously (18) . Cells were washed in Krebs-Ringer solution (0.12 M NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 1.2 mM MgSO4, and 1.2 mM KH2PO4) containing 20 mM HEPES, pH 7.40, 1% (wt/vol) fatty acid-free bovine serum albumin, 100 nM phenylisopropyladenosine, and 0.5 U/ml adenosine deaminase with 2 mM glucose, at 37°C on a shaking water bath. Cells at 20% (vol by vol) were incubated with RBP4 at 50 µg/ml or anti-RBP4 antibodies at 20 µg/ml for 24 h at 37°C, 10% CO2 in the Krebs-Ringer solution mixed with an equal volume of DMEM containing 7% (wt/vol) albumin, 200 nM phenylisopropyladenosine, 20 mM HEPES, 50 UI/ml penicillin, and 50 mg/ml streptomycin, at pH 7.40 (3) . Cells were then washed and transferred to the Krebs-Ringer solution for incubation with or without insulin.

Glucose transport and GLUT4 determination
After transfer of cells to medium without glucose, cells were incubated with indicated concentration of insulin for 15 min, when glucose transport was determined as uptake of 2-deoxy-D-[1-3H]glucose (19) at a final concentration of 50 µM (10 µCi/ml). The cells were incubated for 30 min. It was verified that uptake was linear for at least 30 min.

For determination of amount of GLUT4 at the plasma membrane, plasma membrane sheets were prepared attached to microscopy slides (20) , fixed in 3% paraformaldehyde for 15 min at room temperature and methanol for 2 min at –20°C, blocked with bovine serum albumin, and incubated with monoclonal anti-GLUT4 antibodies. Primary antibodies were detected with fluorescent secondary antibodies (highly cross-adsorbed, Alexa fluor 594, Molecular Probes, Invitrogen, Stockholm, Sweden), and confocal scanning fluorescence microscopy was performed with Nikon D-Eclipse (Nikon, Tokyo, Japan). No labeling was observed in the absence of the primary antibody. Fluorescent intensity was measured with Scion image (Scion, Frederick, MD, USA).

SDS-PAGE and immunoblotting
Cell incubations were terminated by separating cells from medium using centrifugation through dinonylphtalate. To minimize postincubation signaling modifications in the cells and protein modifications, which can occur during immunoprecipitation, the cells were immediately dissolved in SDS and ß-mercaptoethanol with protease and protein phosphatase inhibitors, frozen within 10 s, and thawed in boiling water for further processing (18) . Equal amounts of cells as determined by lipocrit, that is total cell volume, were subjected to SDS-PAGE and immunoblotting (3) . The phosphorylation of IRS1, insulin receptor, and protein kinase B was normalized to the amount of IRS1 protein in each sample.

Data were normalized to percentage of maximal effect, and the effects of insulin on the dose-response curves were fitted to experimental data using the sigmoidal dose-response algorithm of GraphPad Prism 4 (GraphPad Software, Inc., San Diego, CA, USA). Curves were compared statistically with the sigmoidal curve-fitting algorithm in GraphPad Prism 4. When indicated, Student’s t test was used to compare control and treatment groups. The null hypothesis was rejected if P < 0.05.

Isolation of RNA and quantitation of RBP4 mRNA
RNA was prepared from isolated adipocytes by Trizol extraction (Qiazol, Qiagen, Hilden, Germany) and solid-face extraction using RNeasy MinElute Cleanup columns (Qiagen). After reverse transcription (Enhanced Avian First Strand Synthesis Kit, Sigma-Aldrich, St.Louis, MO, USA) with 3 µg of total RNA, real-time quantitative polymerase chain reaction (PCR) was performed using 300 ng cDNA as template with Applied Biosystems’ TaqMan Gene Expression Assays and 7500 Fast Real-Time PCR System (Foster City, CA, USA). Primers and probes for RBP4 and 18S rRNA were obtained from Applied Biosystems. For each primer pair, a standard curve was obtained using serial dilution of human adipose tissue cDNA before mRNA quantitation. 18S rRNA was used as a control to normalize gene expression data. Each sample was analyzed in duplicate, and all analyses were performed at the same time in the same 96-well plate. After 20 s at 95°C, the reaction was run for 45 cycles consisting of denaturation at 95°C for 3 s followed by annealing/extension at 60°C for 30 s. The data were analyzed using 7500 Fast System Sequence Detection System v.1.3.1 (Applied Biosystems).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Effects of RBP4 on the early steps of insulin signaling
To investigate whether RBP4 produces similar signaling defects in the insulin signal chain as we have earlier described in adipocytes from patients with type 2 diabetes (3 , 4) , we examined the effects of RBP4 on isolated adipocytes from the subcutaneous abdominal fat of nondiabetic subjects. Incubation of the adipocytes with RBP4 for 24 h had no effect on the insulin sensitivity for stimulation of insulin receptor autophosphorylation or the maximal insulin-stimulated autophosphorylation (not shown), which is similar to the situation in type 2 diabetes (3) . However, the sensitivity (EC50) to insulin (Fig. 1 A) and maximal effect of insulin (Fig. 1B ) on the downstream phosphorylation of IRS1 on tyrosine were reduced 4- and 2-fold, respectively, in cells incubated with RBP4, thus mimicking adipocytes from patients with type 2 diabetes (3 , 4) .


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		Figure 1. Effect of RBP4 on insulin-induced phosphorylation of IRS1 at tyrosine in human adipocytes. Adipocytes from 5 subjects (age 29–69, average 52 yr; BMI 24–28, average 26 kg/m2) were incubated with (closed circles) or without (open circles) RBP4 at 50 µg/ml for 24 h, when cells were washed and incubated with the indicated concentration of insulin for 10 min. Cells were then subjected to SDS-PAGE and immunoblotting against phosphotyrosine. Phosphorylation of IRS1 is expressed as percentage of maximal insulin-effect in controls and in RBP4-treated cells, respectively (A), or as percentage of maximal insulin-effect in controls (B); mean ± SE (n=5 subjects-independent experiments). EC50 for 2 curves (A) and maximal effects of insulin (B) were statistically different (P<0.05).

Effects of RBP4 on insulin feedback signaling to serine phosphorylation of IRS1
The positive feedback signaling of insulin to enhanced phosphorylation of IRS1 at serine (307) [corresponding to serine (302) in the murine sequence] was inhibited by RBP4 treatment (Fig. 2 ), again mimicking a situation previously shown in adipocytes from patients with type 2 diabetes (4) .


Figure 2
View larger version (12K):
[in this window]
[in a new window]

  		Figure 2. Effects of RBP4 on insulin-induced phosphorylation of IRS1 at serine (307). Adipocytes from 5 subjects (age 29–69, average 52 yr; BMI 24–28, average 26 kg/m2) were incubated with or without, as indicated, 50 µg/ml of RBP4 for 24 h. Cells were then washed and incubated with or without, as indicated, 10 nM insulin for 10 min and subjected to SDS-PAGE and immunoblotting against IRS1-phosphoserine (307). Results are normalized to controls incubated without RBP4 and insulin; mean ± SE (n=5 subjects-independent experiments). The inhibition by RBP4 of insulin-stimulated IRS1-serine (307) phosphorylation was statistically significant (*P<0.05) by Student’s paired t test.

Phosphorylation of the, in the sequence of IRS1, closely situated serine (312) [corresponding to serine (307) in the murine sequence] is in animal cells part of a negative feedback signal from insulin as reviewed in (6 , 7 , 9) . In human adipocytes, basal phosphorylation was low and there was no statistically significant difference between cells from patients with type 2 diabetes compared with cells from nondiabetic subjects (Fig. 3 A). Because the basal level of serine (312) phosphorylation was low, we repeated this comparison with two other groups of diabetic and nondiabetic subjects, again with no difference between diabetic and nondiabetic cells (Fig. 3A ). Insulin stimulation for 30 min (5) increased the phosphorylation on average ~5-fold in cells from both nondiabetic subjects and diabetic patients (Fig. 3B ). RBP4 had no effect on the phosphorylation of IRS1 at serine (312), neither alone nor in combination with insulin (Fig. 3C ).


Figure 3
View larger version (12K):
[in this window]
[in a new window]

  		Figure 3. Phosphorylation of IRS1 at serine (312). A) Basal phosphorylation of serine (312). Experiment 1 = adipocytes from 6 nondiabetic control subjects (age 53–79, average 67 yr; BMI 23–38, average 29 kg/m2) and 6 patients with type 2 diabetes (age 40–73, average 56 yr; BMI 28–47, average 36 kg/m2, HbA1c=4.5–10.2%, average 6.8%); experiment 2 = adipocytes from 7 nondiabetic control subjects (age 56–78, average 65 yr; BMI 22–39, average 27.1 kg/m2) and 7 patients with type 2 diabetes (age 47–70, average 59 yr; BMI 28–48, average 35 kg/m2, HbA1c=5.5–9.7%, average 6.6%). Cells were subjected to SDS-PAGE and immunoblotting against IRS1-phosphoserine (312). Data are mean ± SE; n = 6 subjects (experiment 1) and n = 7 subjects (experiment 2). There was no significant statistical difference between nondiabetic and diabetic. B) Effect of insulin on the phosphorylation at serine (312). Adipocytes from 6 nondiabetic control subjects (age 53–79, average 67 yr; BMI 23–38, average 29 kg/m2) and 6 patients with type 2 diabetes (age 40–73, average 56 yr; BMI 28–47, average 36 kg/m2, HbA1c=4.5–10.2%, average 6.8%) were incubated for 30 min with or without insulin, as indicated. Cells were subjected to SDS-PAGE and immunoblotting against IRS1-phosphoserine (312). Results are normalized to controls incubated without insulin; mean ± SE; n = 6 subjects-independent experiments. There was no significant statistical difference between nondiabetic and diabetic. C) Effect of RBP4 on phosphorylation of serine (312). Adipocytes from the same nondiabetic control subjects as in B were incubated with or without 50 µg/ml RBP4 for 24 h, washed, and then incubated with 10 nM insulin for 30 min, as indicated. Cells were subjected to SDS-PAGE and immunoblotting against IRS1-phosphoserine (312). Results are normalized to controls incubated without RBP4 and insulin; mean ± SE; n = 6, subjects-independent experiments. Effect of insulin was statistically significant (*P<0.05) by Student’s paired t test.

It has been suggested that the amount of IRS1 is reduced in adipocytes from insulin-resistant patients (21) . However, incubation of human adipocytes with RBP4 for 24 h had no effect on the IRS1/actin ratio in the cells as determined by immunoblotting (P=0.88, Student’s paired t test; n=35 determinations from each group). This is in agreement with findings that injection of RBP4 in mice reduced insulin-stimulated tyrosine phosphorylation of IRS1 in muscle without affecting the total amount of IRS1 in the tissue (13) .

Effects of RBP4 on downstream signaling of IRS1
Despite its effects on IRS1 phosphorylation at serine (307) and at tyrosine residues, the insulin-desensitizing effect of RBP4 was not transmitted further downstream to reduced insulin-sensitivity for protein kinase B phosphorylation (Fig. 4 A) or for stimulating glucose uptake (Fig. 4B ). Indeed, the RBP4 treatment increased basal glucose uptake by 50% and this increase was maintained at all concentrations of insulin, without affecting the insulin sensitivity for stimulation of glucose uptake (Fig. 5 A). This increase in glucose uptake could be explained by a parallel increase in the amount of GLUT4 at the plasma membrane in response to the treatment with RBP4 (Fig. 5B ).


Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Figure 4. Effects of RBP4 on insulin-induced phosphorylation of protein kinase B and on glucose uptake. Adipocytes from 5 subjects (age 29–69, average 52 yr; BMI 24–28, average 26 kg/m2) were incubated with (closed circles) or without (open circles) 50 µg/ml of RBP4 for 24 h. Cells were then washed and incubated with the indicated concentration of insulin for 10 min and subjected to SDS-PAGE and immunoblotting against phospho-protein kinase B (PKB; A) or incubated with insulin for 15 min when glucose uptake was determined as uptake of 2-deoxy-D-[1-3H]glucose (B). Results are expressed as percentage of maximal insulin-effect, mean ± SE (n=5 subjects-independent experiments).


Figure 5
View larger version (18K):
[in this window]
[in a new window]

  		Figure 5. Effects of RBP4 on glucose uptake and GLUT4. Adipocytes were incubated with (closed circles) or without (open circles) 50 µg/ml RBP4 for 24 h when cells were washed. A) Effect of RBP4 on glucose uptake. Cells were incubated with the indicated concentration of insulin for 15 min after which glucose transport was determined as uptake of 2-deoxy-D-[1-3H]glucose. Glucose transport is expressed as nanomoles 2-deoxy-glucose per minute per liter of packed cell volume; mean ± SE; n = 5 subjects-independent experiments (age 28–61, average 51 yr; BMI 24–28, average 26 kg/m2). Stimulation by RBP4 of basal glucose uptake was statistically significant (P<0.05) by Student’s paired t test. B) Effect of RBP4 on GLUT4 at plasma membrane. Cells were incubated with 10 nM insulin for 15 min when the amount of GLUT4 protein at the plasma membrane was determined by immunofluorescence confocal microscopy of plasma membrane sheets. 20–30 membrane sheets were analyzed for each subject, with or without RBP4-treatment, and the average intensity per area determined. Data are mean ± SE; n = 4 subjects-independent experiments (age 34–61, average 48 yr; BMI 19–26, average 22 kg/m2). RBP4 treatment produced a significant difference from nontreated control group (*P<0.05) by Student’s paired t test.

Interestingly, the insulin sensitivity for downstream signaling to MAP kinases ERK1/2 phosphorylation was reduced 4-fold by the RBP4 treatment (Fig. 6 A). This faithfully copied the situation in cells from patients with type 2 diabetes, compared with control cells from nondiabetic individuals (Fig. 6B ). There was no significant effect of RBP4 on the maximal phosphorylation of ERK1/2 in response to insulin (not shown).


Figure 6
View larger version (18K):
[in this window]
[in a new window]

  		Figure 6. Effect of RBP4 on insulin-induced phosphorylation of MAP kinase ERK1/2 and phosphorylation of ERK1/2 in cells from nondiabetic subjects vs. patients with type 2 diabetes. A) Effect of RBP4. Adipocytes from 4 subjects (age 29–65, average 48 yr; BMI 25–28, average 26 kg/m2) were incubated with (closed circles) or without (open circles) 50 µg/ml of RBP4 for 24 h. B) Nondiabetic vs. diabetic. Adipocytes from 12 nondiabetic control subjects (age 29–77, average 58 yr; BMI 22–39, average 27 kg/m2; open circles) and 13 patients with type 2 diabetes (age 48–76, average 61 yr; BMI 28–48, average 36 kg/m2, HbA1c=6.3–10.2%, average 7.7%; closed circles) were prepared. A, B) Cells were then washed and incubated with the indicated concentration of insulin for 10 min and subjected to SDS-PAGE and immunoblotting against phospho-ERK1/2. Results are expressed as percentage of maximal insulin-effect; mean ± SE; n = 4 in A; n = 12 controls and n = 13 diabetics in B (n=subjects-independent experiments). EC50 for 2 curves was statistically significantly different (P<0.05) in both A and B.

An autocrine effect of RBP4 on IRS1 serine (307) and ERK1/2 phosphorylation in adipocytes from patients with type 2 diabetes
As RBP4 has been found to be secreted from adipocytes, we posited that RBP4 may act in an autocrine or paracrine fashion in the adipose tissue. It is in such a case possible that adipocytes from patients with type 2 diabetes release RBP4 to induce insulin resistance locally as well as in other more distant tissues. To test this possibility, we used monoclonal antibodies against RBP4, which have earlier been found to block RBP4 interaction with, and delivery of retinol to, HEK293A cells (16 , 17) . We isolated adipocytes from patients with type 2 diabetes and incubated the cells with the anti-RBP4 antibodies for 24 h, when we analyzed the effect of insulin on the phosphorylation of IRS1 at serine (307). Interestingly, the anti-RBP4 antibodies restored the ability of insulin to elicit the phosphorylation of IRS1 at serine (307) (Fig. 7 ), evidently by sequestering endogenously produced and to the medium released RBP4. The rescued phosphorylation of IRS1 at serine (307) was translated into an enhanced sensitivity to insulin for phosphorylation of the downstream MAP kinase ERK1/2 (Fig. 8 A). However, the anti-RBP4 antibody treatment did not improve the insulin sensitivity of glucose uptake (Fig. 8B ) nor did it affect the basal glucose uptake (not shown) in the cells from patients with diabetes.


Figure 7
View larger version (12K):
[in this window]
[in a new window]

  		Figure 7. Anti-RBP4 antibody restoration of insulin-induced phosphorylation of IRS1 at serine (307) in type 2 diabetes. Adipocytes from 5 subjects with type 2 diabetes (age 51–76, average 63 yr; BMI 29–49, average 38 kg/m2, HbA1c=6.0–10.2%, average 7.6%) were incubated with or without, as indicated, 20 µg/ml of anti-RBP4 antibodies ({alpha}-RBP4) for 24 h. Cells were then washed and incubated with or without, as indicated, 10 nM insulin for 10 min and subjected to SDS-PAGE and immunoblotting against IRS1-phosphoserine (307). Results are normalized to controls incubated without anti-RBP4 antibodies and insulin; mean ± SE (n=5 subjects-independent experiments). Effect of insulin on cells treated with anti-RBP4 antibodies was significant (*P<0.05) by Student’s paired t test.


Figure 8
View larger version (19K):
[in this window]
[in a new window]

  		Figure 8. Effect of anti-RBP4 antibody on insulin signaling downstream of IRS1 in adipocytes from patients with type 2 diabetes. A) Effect of anti-RBP4 antibody on ERK1/2 phosphorylation. Adipocytes obtained from 7 subjects with type 2 diabetes (age 40–76, average 59 yr; BMI 29–49, average 40 kg/m2, HbA1c=6.0–10.2%, average 8.0%) were incubated with (filled symbols) or without (open symbols) 20 µg/ml of anti-RBP4 antibodies for 24 h, when cells were incubated with the indicated concentration of insulin for 10 min. Cells were then subjected to SDS-PAGE and immunoblotting against phospho-ERK1/2. Results are expressed as percentage of maximal insulin-effect; mean ± SE (n=7 subjects-independent experiments). EC50 for 2 curves were significantly different (P<0.05). B) Effect of anti-RBP4 antibody on glucose transport in adipocytes from patients with type 2 diabetes. Adipocytes from 4 subjects with type 2 diabetes (age 51–76, average 62 yr; BMI 39–49, average 43 kg/m2, HbA1c=6.1–8.2%, average 7.2%) were incubated with (filled symbols) or without (open symbols) 20 µg/ml of anti-RBP4 antibodies for 24 h, when cells were incubated with the indicated concentration of insulin for 15 min. Glucose transport was then determined as uptake of 2-deoxy-D-[1-3H]glucose. Results are percentage of maximal insulin-effect; mean ± SE (n=4 subjects-independent experiments).

Cellular levels of RBP4 protein and mRNA
Release of RBP4 by the adipocytes, from patients with type 2 diabetes, was corroborated by our finding that human fat cells contained RBP4 and that adipocytes from patients with diabetes or obese subjects had significantly less intracellular RBP4 protein (Fig. 9 A). In contrast, there was no correlation between the levels of RBP4 mRNA in the isolated adipocytes, as determined by real-time PCR, and obesity or BMI (Fig. 9B ).


Figure 9
View larger version (15K):
[in this window]
[in a new window]

  		Figure 9. Levels of RBP4 in human adipocytes. A) Levels of RBP4 protein. Equal amounts of adipocytes from low BMI (24±1 kg/m2, n=4 subjects), high BMI (34±2 kg/m2, n=3 subjects) nondiabetic subjects, or from patients with type 2 diabetes (BMI=34±2 kg/m2, n=7 subjects) were subjected to SDS-PAGE and immunoblotting against RBP4 and actin. Amount of RBP4 protein in cells was normalized to amount of actin in each sample and presented as mean± SE. By Student’s t test, amount of RBP4 in cells from both BMI > 30 and diabetic groups was significantly different from the BMI < 30 group (*P<0.05). B) Levels of RBP4 mRNA. Total RNA was extracted from isolated adipocytes from 12 nondiabetic individuals (age 22–79, average 55 yr; open symbols) and 2 patients with type 2 diabetes (age 40 and 47 yr; HbA1c 6.1 and 10.2%; closed symbols). Amount of RBP4 mRNA was determined by RT-PCR.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The effects of exogenous RBP4 on primary human adipocytes faithfully reproduced the characteristics of the early steps of insulin signaling in cells from patients with insulin resistance and type 2 diabetes: insulin receptor autophosphorylation, insulin receptor phosphorylation of IRS1 at tyrosine residues, and insulin-induced feedback phosphorylation of IRS1 at serine (307). Downstream signaling to phosphorylation of MAP kinase ERK1/2 similarly became insulin resistant as in type 2 diabetes. However, the downstream signaling to metabolic effects via protein kinase B to glucose transport did not correspond to the diabetic state after this RBP4 treatment. We have in this study recruited subjects consecutively as they were scheduled for elective surgery. The only exclusion criterion from nondiabetic controls was a diabetes diagnosis. The inclusion criteria for patients with type 2 diabetes were that subjects should be clinically diagnosed with type 2 diabetes and exhibit the metabolic syndrome. This means that both our control subjects and patients with diabetes are not selected for anything but the presence or absence of the disease. Hence, there will be some obese insulin-resistant subjects in the nondiabetic control groups, but this also allows for a wide significance of our findings.

In human adipocytes, the insulin-induced phosphorylation of IRS1 at serine (307) [corresponding to serine (302) in the murine sequence] appears to be part of a positive feedback loop, which increases the state of phosphorylation of IRS1 at tyrosine in response to insulin and hence sensitizes the further downstream signaling of IRS1 to insulin (4 , 5) . The inhibitor of mTOR, rapamycin, blocks the insulin-induced phosphorylation of serine (307), implicating activation of mTOR by insulin in this positive feedback signaling (4) . A function of serine (307) phosphorylation may therefore be to integrate nutrient sensing with insulin control in adipocytes (22) . This is interesting considering that RBP4 is being produced in increased amounts in cells depleted of GLUT4 with impaired glucose uptake (13) .

Transfection of 32D cells with IRS1-serine (302) [corresponding to serine (307) in the human sequence] mutated to alanine to block that serine phosphorylation has been reported to inhibit insulin-stimulated IRS1 tyrosine phosphorylation, interaction with PI3-kinase, p70S6-kinase phosphorylation, and synthesis of DNA, while not affecting insulin stimulation of protein kinase B (22) . Indeed, herein RBP4 treatment of human adipocytes decreased the insulin sensitivity for MAP kinase ERK1/2 phosphorylation, in correspondence with the effects of RBP4 on IRS1 serine (307) and tyrosine phosphorylation. The dissociation between activation of IRS1/PI3-kinase and the downstream activation of protein kinase B/glucose transport is enigmatic but has been reported also, e.g., in skeletal muscle of diabetic patients (23) and after over-expression of the PH and PTB domains of IRS1 in 3T3-L1 adipocytes (24) . It is possible that phosphorylation of different tyrosine residues in IRS1 may transmit different signals and be differently affected by the serine (307) phosphorylation. An alternative explanation could also be the existence of different pools of IRS1 that are associated with metabolic and mitogenic signaling of insulin and that these pools could be differentially affected by RBP4.

Phosphorylation of IRS1 at serine (312) [corresponding to serine (307) in the murine sequence] has been linked to insulin resistance in animal models, reviewed in (6 , 7 , 9) . We have previously reported that insulin at pathologically high concentrations (EC50=3 nM) induces phosphorylation of IRS1 at serine (312) in human adipocytes and that this corresponds to a reduced level of IRS1 phosphorylation at tyrosine (5) , suggesting that phosphorylation of serine (312) is part of a negative feedback loop. The findings herein that the phosphorylation of serine (312) was not significantly elevated in the diabetic state, that insulin had the same effect in nondiabetic and diabetic cells, and that RBP4 had no effect on basal or insulin-stimulated phosphorylation indicate that phosphorylation of serine (312) is of less importance in type 2 diabetes of humans, as well as in the mechanisms of action of RBP4.

Impaired phosphorylation of protein kinase B and impaired glucose uptake in response to insulin are features of the diabetic state that were not reproduced by RBP4 treatment. This could indicate that RBP4 is not a sufficient physiological inducer of insulin resistance of type 2 diabetes in fat cells. It will be important to examine at the molecular level the effects of RBP4 also in other tissues, primarily skeletal muscle. It is possible that RBP4 affects other aspects of insulin signaling or control of glucose uptake in isolated adipocytes that counteract the insulin resistance in the signaling to stimulate glucose transport, the net effect of which is unchanged insulin sensitivity to enhancement of glucose uptake. We found that RBP4 significantly increased the amount of GLUT4 at the plasma membrane and thereby increased basal glucose uptake. This translocation of GLUT4 was evidently independent of the insulin signal pathway as glucose uptake in response to insulin was additive to the RBP4-increased basal uptake. It is also interesting to consider that many patients with diabetes display a phenotype of normal or near normal fasting glucose uptake but reproducibly have postprandial hyperglycemia.

As for a role of RBP4 to induce an insulin-resistant state that includes metabolic effects of insulin, like glucose transport, the cells were in this study subjected to RBP4 for 24 h and this may not be long enough to elicit full-blown insulin resistance. This notion is supported by the requirement to treat animals with repeated injections of RBP4 for 9 days to induce an insulin-resistant state with glucose intolerance (13) and, of course, in a physiological situation obesity-related insulin resistance and diabetes normally develop over many years. It will be important to study long-term effects of RBP4 treatment on adipocytes.

We report that the impaired phosphorylation of IRS1 at serine (307) in response to insulin and the downstream resistance to insulin for affecting MAP kinase ERK1/2 were normalized after treatment of adipocytes from patients with type 2 diabetes with antibodies against RBP4. This suggests that RBP4 is released from adipocytes obtained from diabetic patients and that RBP4 thus has a role in inducing or maintaining the diabetic state in adipocytes. An increased release of RBP4 from diabetic adipocytes was indeed suggested by the lower levels of RBP4 in adipocytes from patients with diabetes or obese subjects compared with lean, nondiabetic subjects, as earlier findings from human hepatoma HepG2 cells indicate less intracellular RBP4 under conditions of stimulated release of the protein (25) . The expression of RBP4 mRNA in isolated adipocytes, in contrast to the reduced levels of RBP4 protein, was unrelated to obesity. It has even been reported that RBP4 mRNA is slightly down-regulated in whole adipose tissue extracts from obese vs. lean women (14) . This indicates that obesity and diabetes are associated with a higher release rate of RBP4, rather than increased gene activity. It will be important to directly compare the release of RBP4 by adipocytes from patients with type 2 diabetes compared with nondiabetic and lean subjects. However, by culturing and differentiating human preadipocytes it has been demonstrated that human fat cells release RBP4 (14) .

Together with earlier findings of correlations of RBP4 with aspects of insulin resistance in both animals and human beings the findings reported herein support a role of RBP4 in the pathogenesis of insulin resistance and type 2 diabetes. It should not be a surprise though if a single mechanism or agent is unable to explain all aspects of the insulin-resistant state and type 2 diabetes.

We thank P. Kjolhede (Department of Obstetrics and Gynecology), P. Sandström (Department of Surgery), G. Andreescu, B. Lönnberg, and S. Halili (Department of Obstetrics and Gynecology, University Hospital, Linköping, Sweden) for supplying biopsies of adipose tissue and Östergötland County Council, Novo Nordisk Foundation, Swedish Diabetes Association, and Swedish Research Council for financial support. The authors declare no conflicting interests.


   FOOTNOTES
 
1 These authors contributed equally to this work.

Received for publication January 25, 2007. Accepted for publication May 17, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Saltiel, A. R., Kahn, C. R. (2001) Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414,799-806[CrossRef][Medline]
  2. Stenkula, K. G., Said, L., Karlsson, M., Thorn, H., Kjolhede, P., Gustavsson, J., Söderström, M., Strålfors, P., Nystrom, F. H. (2004) Expression of a mutant IRS inhibits metabolic and mitogenic effects of insulin in human adipocytes. Mol. Cell. Endocrinol. 221,1-8[CrossRef][Medline]
  3. Danielsson, A., Öst, A., Lystedt, E., Kjolhede, P., Gustavsson, J., Nystrom, F. H., Strålfors, P. (2005) Insulin resistance in human adipocytes downstream of IRS1 after surgical cell isolation, but at the level of phosphorylation of IRS1 in type 2 diabetes. FEBS J. 272,141-151[CrossRef][Medline]
  4. Danielsson, A., Öst, A., Nystrom, F. H., Strålfors, P. (2005) Attenuation of insulin-stimulated insulin receptor substrate-1 serine 307 phosphorylation in insulin resistance of type 2 diabetes. J. Biol. Chem. 280,34389-34392[Abstract/Free Full Text]
  5. Danielsson, A., Nystrom, F. H., Strålfors, P. (2006) Phosphorylation of IRS1 at serine 307 and serine 312 in response to insulin in human adipocytes. Biochem. Biophys. Res. Commun. 342,1183-1187[CrossRef][Medline]
  6. Zick, Y. (2004) Uncoupling insulin signalling by serine/threonine phosphorylation: a molecular basis for insulin resitance. Biochem. Soc. Trans. 32,812-816[CrossRef][Medline]
  7. Pirola, L., Johnston, A. M., VanObberghen, E. (2004) Modulation of insulin action. Diabetologia 47,170-184[CrossRef][Medline]
  8. Harrington, L. S., Findlay, G. M., Lamb, R. F. (2005) Restraining PI3K: mTOR signalling goes back to the membrane. Trends. Biochem. Sci. 30,35-42[CrossRef][Medline]
  9. Gual, P., LeMarchand-Brustel, Y., Tanti, J.-F. (2005) Positive and negative regulation of insulin signaling through IRS-1 phosphorylation. Biochemie. 87,99-109[Medline]
  10. Carlson, C. J., Koterski, S., Sciotti, R. J., Poccard, G. B., Rondinone, C. M. (2003) Enhanced basal activation of mitogen-activated protein kinases in adipocytes from type 2 diabets. Potential role of p38 in the down-regulation of GLUT4 expression. Diabetes 52,634-641[Abstract/Free Full Text]
  11. Krook, A. M. B. jörnholm, Galuska, D., Jiang, X. J., Fahlman, R., Myers, M. G., Wallberg-Henriksson, H., Zierath, J. R. (2000) Characterization of signal transduction and glucose transport in skeletal muscle from type 2 diabetic patients. Diabetes 49,284-292[Abstract]
  12. Abel, E. D., Peroni, O., Kim, J. K., Kim, Y. B., Boss, O., Hadro, E., Minnemann, T., Shulman, G. I., Kahn, B. B. (2001) Adipose-selective targeting of the GLUT4 gene impairs insulin action in muscle and liver. Nature 409,729-733[CrossRef][Medline]
  13. Yang, Q., Graham, T. E., Mody, N., Preitner, F., Peroni, O. D., Zabolotny, J. M., Kotani, K., Quadro, L., Kahn, B. B. (2005) Serum retinol binding protein 4 contributes to insulin resistance in obesity and type 2 diabetes. Nature 436,356-362[CrossRef][Medline]
  14. Janke, J., Engeli, S., Boschmann, M., Adams, F., Böhnke, J., Luft, F. C., Sharma, A. M., Jordan, J. (2006) Retionol-binding protein 4 in human obesity. Diabetes 55,2805-2810[Abstract/Free Full Text]
  15. Graham, T. E., Yang, Q., Bluher, M., Hammarstedt, A., Ciaraldi, T. P., Henry, R. R., Wason, C. J., Oberbach, A., Jansson, P.-A., Smith, U., et al (2006) Retinol-binding protein 4 and insulin resistance in lean, obese, and diabetic subjects. N. Engl. J. Med. 354,2552-2563[Abstract/Free Full Text]
  16. Melhus, H., Båvik, C.-O., Rask, L., Peterson, P. A., Eriksson, U. (1995) Epitope mapping of a monoclonal antibody that blocks the binding of plasma retinol-binding protein to its receptor. Biochem. Biophys. Res. Commun. 210,105-112[CrossRef][Medline]
  17. Lidén, M., Eriksson, U. (2005) Development of a versatile reporter assay for studies of retinol uptake and metabolism in vivo. Exp. Cell Res. 310,401-408[CrossRef][Medline]
  18. Strålfors, P., Honnor, R. C. (1989) Insulin-induced dephosphorylation of hormone-sensitive lipase. Correlation with lipolysis and cAMP-dependent protein kinase activity. Eur. J. Biochem. 182,379-385[Medline]
  19. Frost, S. C., Kohanski, R. A., Lane, M. D. (1987) Effect of phenylarsine oxide on insulin-dependent protein phosphorylation and glucose transport in 3T3–L1 adipocytes. J. Biol. Chem. 262,9872-9876[Abstract/Free Full Text]
  20. Thorn, H., Stenkula, K. G., Karlsson, M., Örtegren, U., Nystrom, F. H., Gustavsson, J., Strålfors, P. (2003) Cell surface orifices of caveolae and localization of caveolin to the necks of caveolae in adipocytes. Mol. Biol. Cell. 14,3967-3976[Abstract/Free Full Text]
  21. Carvalho, E., Jansson, P., Axelsen, M., Eriksson, J., Huang, X., Groop, L., Rondinone, C., Sjöström, L., Smith, U. (1999) Low cellular IRS1 gene and protein expression predict insulin resistance and NIDDM. FASEB J. 13,2173-2178[Abstract/Free Full Text]
  22. Giraud, J., Leshan, R., Lee, Y.-H., White, M. F. (2004) Nutrient-dependent and insulin-stimulated phosphorylation of insulin receptor substrate-1 on serine 302 correlates with increased insulin signaling. J. Biol. Chem. 279,3447-3454[Abstract/Free Full Text]
  23. Kim, Y.-B., Kotani, K., Ciaraldi, T. P., Henry, R. R., Kahn, B. B. (2003) Insulin-stimulated protein kinase C{lambda}/{zeta} activity is reduced in skeletal muscle of humans with obesity and type 2 diabetes. Diabetes 52,1935-1942[Abstract/Free Full Text]
  24. Sharma, P. M., Egawa, K., Gustafson, T. A., Martin, J. L., Olefsky, J. M. (1997) Adenovirus-mediated overexpression of IRS-1 interacting domains abolishes insulin-stimulated mitogenesis without affecting glucose transport in 3T3–L1 adipocytes. Mol. Cell. Biol. 17,7386-7397[Abstract]
  25. Marinari, L., Lenich, C. M., Ross, A. C. (1987) Production and secretion of retinol-binding protein by a human hepatoma cell line, HepG2. J. Lipid Res. 28,941-948[Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.07-8173comv1
21/13/3696    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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 reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Öst, A.
Right arrow Articles by Strålfors, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Öst, A.
Right arrow Articles by Strålfors, P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS