Sie sind auf Seite 1von 7

0013-7227/04/$15.00/0 Printed in U.S.A.

Endocrinology 145(3):1144 1150 Copyright 2004 by The Endocrine Society doi: 10.1210/en.2003-1297

Human Placental Growth Hormone Increases Expression of the P85 Regulatory Unit of Phosphatidylinositol 3Kinase and Triggers Severe Insulin Resistance in Skeletal Muscle
LINDA A. BARBOUR, JIANHUA SHAO, LIPING QIAO, WAYNE LEITNER, MARIANNE ANDERSON, JACOB E. FRIEDMAN, AND BORIS DRAZNIN
Departments of Medicine (L.A.B., B.D.), Obstetrics and Gynecology (L.A.B.), and Pediatrics (J.S., L.Q., M.A., J.E.F.), University of Colorado Health Sciences Center, and Research Service of the Denver Veterans Affairs Medical Center (W.L., B.D.), Denver, Colorado 80262
The insulin resistance of normal pregnancy is necessary to divert fuels to the fetus to meet fetal growth demands and is mediated by placental hormones. We recently demonstrated that human placental GH (hPGH) can trigger severe insulin resistance in transgenic (TG) mice. In this study we sought to elucidate the cellular mechanisms by which hPGH interferes with insulin signaling in muscle in TG mice. Insulin-stimulated GLUT-4 translocation to the plasma membrane (PM) was reduced in the TG compared with wild-type (WT) mice (P 0.05). Insulin receptor (IR) levels were modestly reduced by 19% (P < 0.01) in TG mice, but there were no changes in phosphorylation of IR or IR substrate-1 (IRS-1) between WT and TG mice. A singular finding was a highly significant increase in the p85 regulatory subunit of phosphatidylinositol 3-kinase (PI 3-kinase; P < 0.001), yet a reduced ability of insulin to stimulate IRS-1-associated PI 3-kinase activity (P < 0.05). Although the levels of the p110 catalytic subunit protein of PI 3-kinase and IRS-1 were unchanged in the TG mice, insulins ability to stimulate p110 association with IRS-1 was markedly reduced (P < 0.0001). We demonstrate a unique mechanism of insulin resistance and suggest that hPGH may contribute to the insulin resistance of normal pregnancy by increasing the expression of the p85 monomer, which competes in a dominant negative fashion with the p85-p110 heterodimer for binding to IRS-1 protein. (Endocrinology 145: 1144 1150, 2004)

HE INSULIN resistance of normal pregnancy is a critical physiological adaptation designed to limit maternal glucose uptake to ensure that an adequate supply of nutrients is shunted to the growing fetus. Placental hormones reprogram mothers physiology to become insulin resistant, and although it has been postulated that human placental lactogen (hPL) is the major insulin resistance hormone of pregnancy, this idea has only minor experimental support. hPL displays both insulin-like and antiinsulin effects, and its main effect is probably directed at the induction of growth of pancreatic islets during pregnancy and stimulation of insulin secretion (1 4). Recently, we demonstrated that human placental GH (hPGH), expressed in transgenic (TG) mice at levels comparable to those in the third trimester of human pregnancy, can cause severe total body insulin resistance (5). hPGH is a product of the hPGH variant gene and member of the GH family that differs from pituitary GH by 13 amino acids (6, 7). It is not regulated by GHRH and is secreted tonically from the placenta throughout gestation. It has the same affinity for the GH receptor and almost completely replaces pituitary GH in the maternal circulation by 20 wk of pregnancy (7).
Abbreviations: hPGH, Human placental GH; hPL, human placental lactogen; IR, insulin receptor; IRS-1, insulin receptor substrate-1; PI 3-kinase, phosphatidylinositol 3-kinase; PM, plasma membrane; TBS-T, Tris-buffered saline-Tween 20; TG, transgenic; WT, wild-type. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

hPGH does not cross the placenta and appears to regulate the maternal levels of IGF-I that parallels the increasing levels of hPGH after 20 wk of pregnancy (3). We previously characterized TG mice that overexpressed hPGH at levels similar to those measured in the third trimester of human pregnancy (5). These mice displayed severe insulin resistance with overt hyperinsulinemia to maintain normoglycemia. The insulin levels in TG mice were 4- to 7-fold greater than those in their wild-type (WT) littermates (fasting and postprandial, respectively), and glucose disposal, as measured by an insulin challenge test, was markedly reduced. The present study was designed to evaluate the mechanisms by which hPGH can cause insulin resistance in these TG mice overexpressing hPGH. We examined the expression, phosphorylation, and activity of several signaling intermediates involved in the metabolic action of insulin in skeletal muscle as well as the effect of hPGH on the insulinstimulated translocation of GLUT-4 to the plasma membrane. Our results indicate that hPGH mediates insulin resistance through a unique action by specifically increasing the expression of the p85 regulatory unit of phosphatidylinositol 3-kinase (PI 3-kinase), resulting in a marked reduction in insulin receptor substrate-1 (IRS-1)-associated PI 3kinase activity.
Materials and Methods Animals
TG mice that overexpress hPGH (hereafter, TG) have been described previously (5, 8). The transgene is driven by the mouse metallothionein-1

1144

Barbour et al. Placental GH and Insulin Resistance

Endocrinology, March 2004, 145(3):1144 1150 1145

promoter and is expressed primarily in the liver, small intestine, and kidney during fetal development and throughout postnatal life. The nontransgenic WT littermates have identical genetic backgrounds, except for the absence of the hPGH transgene. We found the TG mice to be extremely insulin resistant when examined by glucose and insulin challenge tests (5) (Table 1). Free fatty acid levels were not different in the two groups of mice. Furthermore, we found these animals to weigh nearly twice as much as their littermates, but their percent fat mass was slightly, but statistically, lower (15%) than that of WT mice. The hPGH levels were 92.25 ng/ml in TG mice compared with undetectable in the WT mice, and IGF-I levels were increased in TG mice (5). Mean fasting insulin levels were 1.57 0.22 compared with 0.38 0.07 ng/ml (P 0.001) in TG and WT mice, respectively, and both groups were normoglycemic at 16 wk of age. Four to six TG and four to six WT mice were used in each comparison. The TG and WT mice were studied at 16 wk and were fed a normal mouse diet ad libitum. The guidelines for the care and use of the animals were approved by the animal research laboratory at University of Colorado Health Sciences Center.

brane was washed three times with TBS-T and probed with a monoclonal anti-IR (1:1000 dilution in TBS-T), anti-IRS-1 (1:1000 dilution in TBS-T), anti-p85 antibody (1:2000 dilution in TBS-T), or anti-p110 antibody (1:500 dilution in TBS-T). Transfer and washing conditions were described previously (9). The bands were visualized with enhanced chemiluminescence and exposed to Kodak Biomax films (Eastman Kodak Co., Rochester, NY). The specific bands were quantitated using a Gel-Doc density scanner and Quantity One software (Bio-Rad Laboratories).

Phosphorylation of IR and IRS-1


To determine the Ser307 phosphorylation of IRS-1, 75-g protein (using BSA as a standard) samples were subjected to 7% SDS-PAGE. After transferring and blocking, the membrane was incubated with anti-IRS1-Ser307 antibody (1:1000 in TBS-T with 1% BSA overnight at 4 C). The membrane was detected as described above. Tyrosine phosphorylation of IR was determined by incubating 500 mg protein from the muscle homogenate overnight at 4 C with antiphosphotyrosine antibody (5 mg AB/8 mg protein) in 1 ml immunoprecipitation buffer containing 2% Triton X-100, 300 mmol/liter NaCl, 20 mmol/liter Tris-HCl, 2 mmol/ liter EDTA, 1 mmol/liter EGTA, 0.4 mmol phenylmethylsulfonylfluoride, 0.4 mmol/liter sodium vanadate, and 1% Nonidet P-40. After immunoprecipitation, the samples were mixed with 50 ml protein ASepharose (10% solution) for 4 h at 4 C, and the immunoprecipitate was washed in 1 ml immunoprecipitation buffer, followed by centrifugation at 500 g for 1 min at 4 C; this was repeated four times. The washed precipitate was mixed with Laemmli sample buffer (50 ml), boiled for 5 min, centrifuged for 5 min at 500 g, and subjected to 7% SDS-PAGE. After transferring and blocking, the membrane was incubated with anti-IR or anti-IRS-1 antibodies. Bands were visualized by the previously described methodology.

Materials
Regular insulin was purchased from Novo Nordisk (Princeton, NJ). BSA and the protease inhibitors, aprotinin and leupeptin, were purchased from Roche (Indianapolis, IN). Polyclonal antibodies to the insulin receptor (IR) and p110 were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies to IRS-1, agarose-conjugated IRS-1, and p85 were purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). GLUT-4 antibody was obtained from Chemicon International (Temecula, CA). Ser307-IRS-1 antibody was supplied by Dr. Morris White (Joslin Laboratory, Boston, MA). Secondary horseradish peroxidase-conjugated antibody, protein A-Sepharose, and chemiluminescence reagents (ECL kit) were obtained from Amersham Pharmacia Biotech (Piscataway, NJ). AG resins, polyvinylidene difluoride membranes, PAGE gel equipment, and protein assay kits were purchased from Bio-Rad Laboratories (Hercules, CA). [-32P]ATP was obtained from NEN (Boston, MA).

Association of IRS-1 with p85 and p110


Immunoprecipitation of IRS-1 was performed by washing anti-IRS-1 agarose-conjugated beads (Upstate Biotechnology, Inc.) with PBS three times for 1 min each time at 14K rpm. A 400-g lysate of protein was added to anti-IRS-1 agarose beads (10 g anti-IRS-1), and the samples were rotated overnight at 4 C. The next morning, the samples were microfuged for 10 min at 4000 rpm at 4 C. The supernatant was removed, and the pelleted beads were washed with 600 ml ice-cold PBS for 2 min and microfuged three times. Pellets were speed-vacuumed to dryness, and 20 l 2 Laemmli buffer were added to each sample. After boiling 3 min, samples were subjected to 7% SDS-PAGE. After transfer, the membrane was blocked overnight at 4 C with 5% milk in TBST. The membrane was rinsed three times for 20 min each time with TBST. A 20-ml quantity of a 1:1000 dilution of p85, p110, or IRS-1 in 5% milk/ TBST was rocked overnight at 4 C, and the bands were visualized as described above.

Acute insulin stimulation in vivo and tissue collection


The mice were fasted for 6 h and anesthetized with ketamine (150 mg/kg) and acepromazine (5 mg/kg), abdominal cavities were opened, and portal veins were exposed. Approximately 300 mg gastrocnemius muscle from one hindlimb were rapidly removed and frozen immediately in liquid nitrogen. An insulin bolus of 10 U/kg body weight was then injected into the portal vein as described previously (9). Five minutes after injection, the gastrocnemius muscle from the opposite limb was excised and frozen immediately. The samples were stored at 80 C until analysis.

Western blotting of IR, IRS-1, p85, and p110


To determine the total protein levels of IR, IRS-1, p85, and p110, 75 g muscle tissue homogenate protein were treated with Laemmli sample buffer, boiled for 5 min, resolved on a 7% denaturing SDS-PAGE gel, and transferred to polyvinylidene difluoride membranes. Membranes were blocked with 5% nonfat milk (Bio-Rad Laboratories) in Tris-buffered saline-Tween 20 (TBS-T) for 1 h at room temperature. The mem-

IRS-1-associated PI 3-kinase activity


The level of IRS-1-associated PI 3-kinase activity was determined in muscle extracts after immunoprecipitation with IRS-1 antibody overnight at 4 C (400 g muscle protein/4 g antibody), followed by incubation with protein A-Sepharose overnight. The immunoprecipitation complex was spun at 14,000 g for 10 min, followed by washing three

TABLE 1. Body composition and insulin and glucose levels during insulin and glucose tolerance tests in wild-type and transgenic mice overexpressing hPGH
WT Transgenic P value

Weight (g) Fat mass (%) Bone mineral density (g/cm2) Fasting insulin (ng/ml) Insulina 30 min after ip glucose Fasting glucose (mg/dl) Glucosea 30 min after ip insulin

25.3 1.02 17.1 0.88 0.582 0.0017 0.38 0.07 0.62 0.10 130 4.31 62 4.0

46.8 2.66 14.5 0.54 0.641 0.0006 1.57 0.22 4.17 0.54 123 7.0 97 4.5

0.001 0.05 0.01 0.01 0.01 NS 0.001

Comparison of mean SEM between groups, with two-sample t tests for independent samples. a The glucose dose was 2 g/kg body weight; the ip insulin dose was 0.75 U/kg body weight.

1146

Endocrinology, March 2004, 145(3):1144 1150

Barbour et al. Placental GH and Insulin Resistance

times with isotonic PBS containing 1% Nonidet P-40, twice in 0.5 m LiCl2/100 mm Tris-HCl (pH 7.6), and twice with 10 mm Tris-HCl (pH 7.4), 100 mm NaCl, and 1 mm EDTA. The pellets were resuspended in 50 l of the final wash buffer, and 10 l 100 mm MgCl2 were added along with 10 l of a mixture containing 0.5 mg/liter -phosphatidylinositol (Sigma-Aldrich Corp., St. Louis, MO) in 10 mm Tris/1 mm EGTA, and the tube was sonicated for 20 sec. To start the PI 3-kinase reaction, 10 l ATP mix containing 100 mm MgCl2, 10 mm Tris (pH 7.5), 0.55 mm ATP, and 1 mCi/ml [-32P]ATP were added for 10 min at room temperature. The reaction was stopped with 20 l 8 n HCl, and 5 min later 160 l CHCl3:MeOH were added (1:1). The phases were separated by centrifugation, and the lower organic phase was removed, lyophilized to dryness, and resuspended in 15 l ethanol. Five microliters of the product were then loaded on to a silica gel thin layer chromatography plate precoated in 1% potassium oxalate. The lipids were resolved in CHCl3/ MEOH/H2O/NH4OH (60:47:11.3:2), dried, and visualized by autoradiography. The images were quantified using a Kodak Dynamic phosphorimager, and results (in duplicate) were expressed as the percent stimulation over basal (arbitrary units) relative to the WT mice.

GLUT-4 translocation to the plasma membrane (PM)


Approximately 0.3 g muscle was pulverized in liquid nitrogen and placed in 2 ml ice-cold homogenization buffer [5 ml 0.25 m sucrose, 20 ml 0.5 mm EDTA (pH 8.0), 1 ml 50 mm HEPES (pH 7.7), 40 ml 2 mg/ml aprotinin, 20 ml 5 mg/ml leupeptin, 100 ml 0.2 mm PMSF, 60 ml 3 mm dithiothreitol, and 13.7 ml ddH2O) for each sample studied (four WT and four TG mice). The whole homogenate was transferred to a glass vial and homogenized using a motorized Teflon pestle. Each homogenate sample was then centrifuged for 10 min at 1,200 g, and the supernatant was transferred to a 15-ml conical tube. To further increase the yield of plasma membrane, the pellet was resuspended in 1 ml homogenization buffer, homogenized, and centrifuged a second time at 1,200 g. The supernatant from this second homogenized sample was added to the supernatant in the 15-ml conical tube, and the pellet that contained the nuclei was discarded. The supernatant fractions were centrifuged for 15 min at 9,000 g to remove the lysosomes and peroxisomes. Pellets were discarded, and supernatants were transferred to polypropylene tubes and centrifuged at 25,000 g. The pellet containing the Golgi and sarcoplasmic reticulum was discarded, and the supernatant was transferred to ultracentrifuge tubes and subjected to 100,000 g centrifugation for 90 min to separate the remaining cytosolic fraction and to further enrich the plasma membranes. The pellets containing the plasma membranes were resuspended in 200 ml pellet resuspension buffer [0.25 m sucrose, 50 mm HEPES (pH 7.7), 100 mm KCl, 5 mm MgCl2, and protease inhibitors/dithiothreitol in ddH2O] and frozen at 80 C until analysis. For Western analysis, 75 g plasma membrane protein was subjected to 12% SDS-PAGE using rabbit anti-GLUT-4 (1:1,000 dilution in 5% milk-TBS) according to the methods outlined above. FIG. 1. GLUT-4 translocation to the plasma membrane in WT and TG mice. PM fractions were isolated from the skeletal muscle of WT and TG (hPGH) mice before and after in vivo insulin stimulation by differential centrifugation. The results represent the percent change in the amount of GLUT-4 protein in the PM fraction after insulin stimulation. The amount of GLUT-4 protein present in PM fraction after insulin injection in TG mice was approximately 5-fold less than that in the WT mice. A representative blot is shown in the upper panel. Results are expressed as the mean and SEM of four determinations in each group. *, P 0.05 vs. corresponding TG.

minimal and insignificant change in the PM fraction in the TG mice. The amount of GLUT-4 present in the PM fraction after insulin injection in TG mice was approximately 5-fold less than that in WT mice (P 0.05), confirming insulin resistance in mice expressing hPGH.
Insulin signaling proteins

Statistical analysis
A two-sample t test for independent samples with equal variances were used for all comparisons. The results were represented as the mean sem. P 0.05 was considered significant.

Results GLUT-4 translocation to the PM in skeletal muscle of WT and TG mice

Because the mechanism responsible for translocation of GLUT-4 to the PM appears to involve both signal transduction from IR to IRS-1 as well as PI 3-kinase, we examined this signaling pathway in more detail. In our initial experiments we measured the expression of the IR, IRS-1, p85, and p110 subunits of PI 3-kinase in skeletal muscle of WT and TG mice. The total IR protein levels were modestly, but statistically significantly, decreased by 19% (P 0.01) in TG mice, but IRS-1 levels were not different between the groups of mice (Fig. 2). Most notable was a significant increase in the basal levels of p85 (Fig. 2), a regulatory subunit of the PI 3-kinase (P 0.001). In contrast, p110 levels, the catalytic subunit of PI 3-kinase, were unchanged.
Phosphorylation of IR and IRS-1

Given that our previously reported results (5) demonstrated a marked decrease in glucose disposal during the insulin challenge test in TG mice, in the present investigation we initially determined the ability of insulin to promote GLUT-4 translocation to the PM of skeletal muscle. The total amount of GLUT-4 in muscle homogenates was unchanged in TG compared with WT mice (not shown). However, although insulin promoted an approximately 2-fold increase in the appearance of GLUT-4 in the PM fractions isolated from the skeletal muscle of WT mice (Fig. 1), there was only a

We then assessed basal and insulin-increased tyrosine phosphorylation of IR and IRS-1 protein in the skeletal muscle of WT and TG mice. Insulin stimulated tyrosine phosphorylation of both IR and IRS-1, with the maximal effect of insulin being similar in WT and TG mice (not shown). The level of serine phosphorylation of IRS-1 determined with a site-specific antibody (Ser307-IRS-1) was also unchanged in TG mice (Fig. 3).

Barbour et al. Placental GH and Insulin Resistance

Endocrinology, March 2004, 145(3):1144 1150 1147

In these experiments, IRS-1-associated PI 3-kinase activity was measured in IRS-1 immunoprecipitates 5 min after insulin injection. As expected, insulin significantly increased IRS-1-associated PI 3-kinase activity in the skeletal muscle of WT mice (Fig. 4). In contrast, insulin-stimulated, IRS-1associated PI 3-kinase activity was significantly reduced in TG compared with WT mice (P 0.05).
p85 and p110 association with IRS-1

FIG. 2. Total protein levels of IRS-1, IR, and the p85 and p110 subunits of PI 3-kinase. Equivalent amounts of protein isolated from the muscle of WT and hPGH TG mice were subjected to SDS-PAGE gel and blotted with anti-IRb, anti-IRS-1, and anti-PI 3-kinase p85 subunit and p110 subunit antibodies. A representative blot is shown in the upper panel. The values are the mean SEM of the scanning densitometry values (n 4 6 in each group) expressed in arbitrary units. *, P 0.05; #, P 0.001 (vs. TG mice).

Given the significant decrease in IRS-1-associated PI 3kinase activity in the presence of increased p85 protein, we determined whether insulin could stimulate binding of the PI 3-kinase heterodimer (p85-p110) to IRS-1. In the basal state, the association of p85 with IRS-1 was increased about 2-fold in TG compared with WT mice (Fig. 5A; P 0.001), consistent with increased p85 expression in muscle of the TG mice. However, after insulin stimulation, there was a significant increase in the association of p85 with IRS-1 in the WT mice, but no further increase in TG mice (P 0.001). Similarly, after insulin stimulation, the association of the p110 catalytic unit with IRS-1 increased more than 5-fold in WT mice (P 0.0001), but it did not change in TG mice, suggesting that high levels of p85 monomer block an association of the p85-p110 heterodimer with IRS-1, resulting in lower PI 3-kinase activity (Fig. 5B). To ensure that the amount of IRS-1 was immunoprecipitated equally in WT and TG samples, IRS-1 protein was determined in all IRS-1 immunoprecipitates (Fig. 5C). IRS-1 was equally immunoprecipitated in WT and TG samples.

FIG. 3. Serine phosphorylation of IRS-1 in WT and hPGH TG mice. Serine phosphorylation of IRS-1 was determined by subjecting equivalent amounts of protein to SDS-PAGE and blotted with anti-IRS-1Ser307. A representative blot is shown in the upper panel. The values are the mean SEM of the scanning densitometry values (n 4 6 in each group) expressed in arbitrary units.

IRS-1-associated PI 3-kinase activity

Because the induction of PI 3-kinase activity is critical for insulin-stimulated glucose uptake in skeletal muscle, we next assessed this aspect of insulin signaling in WT and TG mice.

FIG. 4. IRS-1-associated PI 3-kinase activity in WT and TG mice. IRS-1 was immunoprecipitated from muscle extracts obtained before and after in vivo insulin stimulation. PI 3-kinase activity was determined as described in Materials and Methods. The upper panel depicts a representative experiment. The values are the mean SEM of the scanning densitometry values (n 4 in each group) expressed in arbitrary units. *, P 0.05 vs. PI 3-kinase activity in TG mice.

1148

Endocrinology, March 2004, 145(3):1144 1150

Barbour et al. Placental GH and Insulin Resistance

Discussion

FIG. 5. Association of PI 3-kinase subunits with IRS-1 in WT and hPGH TG mice. A, Association of the p85 regulatory subunit with IRS-1; B, sssociation of the p110 catalytic subunit of PI 3-kinase with IRS-1; C, Levels of IRS-1 protein captured in IRS-1 immunoprecipitate in WT and hPGH TG mice. Immunoprecipitation of IRS-1 was performed as described in Materials and Methods, followed by blot-

The remarkable finding of the present study is that the severe insulin resistance in TG mice overexpressing hPGH appears to be due to a specific block of PI 3-kinase activation associated with increased expression of the p85 regulatory subunit of this enzyme in skeletal muscle. The PI 3-kinase signaling pathway is activated by insulin binding to its cell surface receptor with immediate activation of tyrosine kinase activity and phosphorylation of the IR and IRS-1 proteins (10). Although we found a modest decrease in the number of IRs in TG mice, the most likely explanation for this observation is that the hyperinsulinemia present in these animals down-regulates the number of IRs. The fact that tyrosine and serine phosphorylation of IRS-1 was identical in WT and TG mice suggests that the mechanism of the hPGH-induced resistance does not involve impairment at the level of IRS-1. In contrast, a dramatic increase in the amount of p85 subunit of PI 3-kinase drew our attention to this signaling intermediate. PI 3-kinase is composed of a regulatory subunit (p85) and a catalytic subunit (p110), which must form a heterodimer for PI 3-kinase activation to occur (11). This heterodimer then binds to tyrosine-phosphorylated IRS-1 (YXXM motif) using the SH2 domains of the p85 subunit. Here we demonstrate a decreased association of the p85-p110 heterodimer with IRS-1 after insulin stimulation despite an increased basal association of the p85 subunit with IRS-1. Our results suggest that excess p85 monomer can compete with the p85p110 heterodimer for binding to IRS-1 in a dominant negative fashion, thereby causing a decrease in the IRS-1-associated PI 3-kinase activity and a subsequent reduction in GLUT-4 translocation to the plasma membrane. The latter has been clearly shown to be dependent upon normal PI 3-kinase activation (12). Our hypothesis that increased expression of the p85 monomer could exert an inhibitory influence on PI 3-kinase activity is based on densitometry of Western blots, but is also supported by a number of publications in the literature. Terauchi and colleagues (13) first demonstrated that mice with a targeted heterozygous disruption of the gene encoding the p85 regulatory subunit of PI 3-kinase showed increased insulin sensitivity and hypoglycemia due to increased glucose transport in skeletal muscle and adipocytes. Ueki et al. (14) demonstrated a similar phenomenon in L6 myotubes, where increasing the levels of p85 inhibited both phosphotyrosine- and p110-associated PI 3-kinase activities. Mauvais-Jarvis and Kahn (15) also demonstrated that heterozygous deletion of the Pik3r1 gene coding for murine p85 and its splice variants actually improved insulin sensitivity in normal mice as well as in mice made insulin resistant by heterozygous deletions of IR or IRS-1. Mice with a deletion of p85 had lower fasting and postprandial glucose levels and a significant decrease in the incidence of diabetes when crossed with genetically insulin-resistant mice (15). Although our results are also consistent with an inhibitory role
ting with p85, p110, and IRS-1 antibodies. The upper panels depict representative experiments. The values are the mean SEM of the scanning densitometry values (n 4 6 in each group) when expressed in arbitrary units. #, P 0.001 vs. TG; *, P 0.0001 vs. TG.

Barbour et al. Placental GH and Insulin Resistance

Endocrinology, March 2004, 145(3):1144 1150 1149

of p85, future experiments must address potential changes in other isoforms of p85 and p110 as well as the availability of free p85 in relationship to p85-p110 dimers. Insulin-resistant states may modify the expression of these isoforms differentially (16). There are three alternative p85 splice isoforms and four different p110 isoforms that could also be differentially regulated in insulin-resistant states. Any of these could play an additional role in uncoupling the p85 to p110 ratio. However, our finding of reduced PI 3-kinase activation is consistent with an inhibitory role for p85 given that phosphorylation of IRS-1 was unaffected, yet the excess p85 monomer prevented formation of the p85-p110 dimer. Thus, we postulate that a moderate increase in the expression of the p85 monomer resulted in altered stoichiometry of the p85-IRS complex compared with the p85/p110/IRS-1 complex necessary for PI 3-kinase activation. This may similarly apply to the insulin resistance of pregnancy and the insulin resistance induced by the expression of hPGH. Even though the structure and regulation of hPGH differ from those of pituitary GH, both hormones bind the same receptors and could work in a similar fashion. The mechanism by which pituitary GH causes insulin resistance, as seen in acromegaly, has been examined using a TG mouse model (17). TG mice overexpressing pituitary GH demonstrated decreased expression of skeletal muscle IR and decreased tyrosine phosphorylation of IRS-1. Consistent with our data, pituitary GH overexpression was also associated with an increased abundance of the p85 subunit of PI 3-kinase and a decrease in insulin-stimulated PI 3-kinase activity. Similarly, the association of p85 to IRS-1 was increased in the basal state in mice overexpressing pituitary GH. However, unlike WT mice, which showed an increase in the association of p85 to IRS-1 after insulin stimulation, no such increase was noted in TG mice overexpressing pituitary GH (17). Thus, there may be a common mechanism between the action of pituitary and placental GH in triggering insulin resistance in skeletal muscle. The question of why normal tyrosine and serine/threonine phosphorylation of IRS-1 is not accompanied by even further increased association with p85 in TG mice in this and other studies remains enigmatic. However, this may be a function of the chronic hyperinsulinemia, hPGH, or both in these animals that causes increased p85 expression and prevents further recruitment to IRS-1. Protein kinase B (Akt) as well as protein kinase C are downstream markers of PI-3 kinase activation. However, GH has been demonstrated to activate Akt and protein kinase C independently of insulin (18 21). Thus, measurements of Akt activation cannot be used to assess insulin action in TG animals overexpressing hPGH, and the clearest indicator of insulin-induced PI 3-kinase activation is GLUT-4 translocation to the plasma membrane. The insulin resistance of normal pregnancy is an essential physiological adaptation designed to restrict maternal glucose uptake to ensure that adequate nutrients are shunted to the growing fetus. This insulin resistance is thought to be mediated by placental hormones that reprogram the mothers metabolism to divert a sufficient supply of glucose to meet fetal growth demands. Although hPL has classically been postulated to mediate this insulin resistance, hPL appears to play a more central role in stimulating insulin se-

cretion by inducing the growth of pancreatic islets (4). Kirwan and colleagues (22) recently reported that in humans, circulating plasma TNF was produced by the placenta and correlated highly with in vivo insulin resistance during pregnancy, but hPGH was not studied. We demonstrated that hPGH is a likely candidate to mediate the insulin resistance of normal pregnancy (5). We found that TG mice overexpressing hPGH at levels similar to those found in the third trimester of human pregnancy (92 ng/ml) exhibited extreme insulin resistance characterized by fasting and postprandial hyperinsulinemia and decreased insulin-stimulated glucose disposal. These findings were the first to implicate hPGH as a metabolically active hormone capable of inducing the insulin resistance of pregnancy. Together these findings challenge the classic paradigm of hPL as the singular insulin resistance hormone of pregnancy and suggest PGH as a cause of insulin resistance in gestation. Our results agree with observations made by Friedman et al. (23) and Shao et al. (24) who studied the insulin resistance of human pregnancy. Friedman found that the skeletal muscle of healthy pregnant women in their third trimester of pregnancy manifested a nearly 2-fold increase in p85 levels. Although they did not report PI 3-kinase activity, there was a finding of reduced insulin-stimulated glucose transport (23), suggesting that increased p85 may have contributed to the insulin resistance. In conclusion, we demonstrate that hPGH causes severe insulin resistance by specifically increasing the expression of the p85 subunit and subsequently affecting the ability of insulin to stimulate the association of the p85-p110 heterodimer with IRS-1, reducing PI 3-kinase insulin signaling. Although the effect by which hPGH selectively increases the p85 subunit and promotes insulin resistance may also involve the SOCS-1 (suppressors of cytokine signaling) and -6 or Janus kinase-2 pathways, which can cross-talk with insulin signaling (21, 25), the current data allow us to postulate that high levels of the p85 monomers may compete in a dominant negative fashion with the p85-p110 heterodimer

FIG. 6. Schematic of hPGH mechanistic action on the insulin signaling pathway. hPGH specifically increases the p85 monomer, which competitively inhibits the p85-p110 heterodimer from binding to IRS-1, resulting in a decrease in insulin-induced, IRS-1-associated-PI 3-kinase activation and translocation of GLUT-4 to the PM.

1150

Endocrinology, March 2004, 145(3):1144 1150

Barbour et al. Placental GH and Insulin Resistance

for binding to tyrosine-phosphorylated IRS-1 proteins. We demonstrate a unique mechanism of insulin resistance in these mice and suggest that hPGH may contribute to the insulin resistance of normal pregnancy secondary to its effect on p85 expression and its interference with PI 3-kinase activity in skeletal muscle (Fig. 6).
Acknowledgments
We thank Dr. A. Bartke for his critical reading of this manuscript and generous donation of the hPGH transgenic mice. Received September 29, 2003. Accepted November 11, 2003. Address all correspondence and requests for reprints to: Linda A Barbour, M.D., University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Box B-198, Denver, Colorado 80262. E-mail: lynn. barbour@uchsc.edu. This work was supported by NIH Grants 1K23-RR-017496-01 (to L.A.B.) and DK-62155 (to J.E.F.).

References
1. Yamashita H, Shao J, Friedman JE 2000 Physiologic and molecular alterations in carbohydrate metabolism during pregnancy and gestational diabetes. Clin Obstet Gynecol 43:8798 2. Handwerger S, Freemark M 2000 The roles of placental growth hormone and placental lactogen in the regulation of human fetal growth and development. J Pediatr Endocrinol Metab 13:343356 3. Anthony RV, Limesand SW, Fanning MD, Liang R 1998 Placental lactogen and growth hormone. In: Bazer FW, ed. Endocrinology of pregnancy. Totowa, NJ: Humana Press; 461 490 4. Brelje TC, Scharp DW, Lacy PE, Ogren L, Talamantes F, Robertson M, Friesen HG, Sorenson RL 1993 Effect of homologous placental lactogens, prolactins, and growth hormones on islet B-cell division and insulin secretion in rat, mouse, and human islets: implication for placental lactogen regulation of islet function during pregnancy. Endocrinology 132:879 887 5. Barbour LA, Shao J, Qiao L, Pulawa LK, Jensen DR, Bartke A, Garrity M, Draznin B, Friedman JR 2002 Human placental growth hormone causes severe insulin resistance in transgenic mice. Am J Obstet Gynceol 186:512517 6. Evain-Brion D, Alsat E, Igout A, Frankenne F, Hennen G 1994 Placental growth hormone variant: assay and clinical aspects. Acta Paediatr 399(Suppl): 49 51 7. Alsat E, Guibourdenche J, Couturier A, Evain-Brion D 1998 Physiological role of human placental growth hormone. Mol Cell Endocrinol 140:121127 8. Selden RF, Wagner TE, Blethen S, Yun JS, Rowe ME, Goodman HM 1988 Expression of the human growth hormone variant in cultured fibroblasts and transgenic mice. Proc Natl Acad Sci USA 85:8241 8245 9. Shao J, Yamashita H, Qiao L, Draznin B, Friedman JE 2002 Phosphatidylinositol 3-kinase redistribution is associated with skeletal muscle insulin resistance in gestational diabetes. Diabetes 51:19 29

10. Pessin JE, Kahn AH 2002 Insulin regulation of glucose uptake: a complex interplay of intracellular signaling pathways. Diabetologia 45:14751483 11. Le Roith D, Zick Y 2001 Recent advances in our understanding of insulin action and insulin resistance. Diabetes Care 24:588 593 12. Jiang G, Zhang BB 2002 PI 3-kinase and its up- and down-stream modulators as potential targets for the treatment of type II diabetes. Front Biosci 7:d903 d917 13. Terauchi Y, Tsuji Y, Satoh S, Minoura H, Murakami K, Okuno A, Inukai K, Asano T, Kaburagi Y, Ueki K, Nakajima H, HanafusaT, Matsuzawa Y, Sekihara H, Yin Y, Barrett JC, Oda H, Ishikawa T, Akanuma Y, Komuro I, Suziki M, Yamamura K, Kodama T, Suzuki H, Koyasu S, Aizawa S, Tobe K, Fukui Y, Yazaki Y, Kadowaki T 1999 Increased insulin sensitivity and hypoglycaemia in mice lacking the p85 subunit of phosphoinositide 3-kinase. Nat Genet 21:230 235 14. Ueki K, Algenstaedt P, Mauvais-Jarvis F, Kahn CR 2000 Positive and negative regulation of phosphoinositide 3-kinase-dependent signaling pathways by three different gene products of the p85 regulatory subunit. Mol Cell Biol 20:8035 8046 15. Mauvais-Jarvis F, Ueki K, Fruman DA, Hirshman MF, Sakamoto K, Goodyear LJ, Iannacone M, Accili D, Cantely LC, Kahn CR 2002 Reduced expression of the murine p85 subunit of phsophoinositide 3-kinase improves insulin signaling and ameliorates diabetes. J Clin Invest 109:141149 16. Kerouz NJ, Horsch D, Pons S, Kahn CR 1997 Differential regulation of insulin receptor substrates-1 and -2 (IRS-1 and IRS-2) and phosphatidylinositol 3kinase isoforms in liver and muscle of the obese diabetic (ob/ob) mouse. J Clin Invest 100:3164 3172 17. Dominici FP, Cifone D, Bartke A, Turyn D 1999 Alterations in the early steps of the insulin-signaling system in skeletal muscle of GH-transgenic mice. Am J Physiol 277:E447E454 18. Sakaue H, Ogawa, Takata M, Kuroda S, Kotani K, Matsumoto M, Sakue M, Nishio S, Ueno H, Kasuga M 1997 Phosphoinositide 3-kinase is required for insulin-induced but not for growth hormone- or hyperosmolarity-induced glucose uptake in 3T3L1 adipocytes. Mol Endocrinol 11:15521562 19. Takano A, Haruta T, Iwata M, Usui I, Uno T, Kawahara J, Ueno E, Sasaoka T, Kobayashi M 2001 Growth hormone induces cellular insulin resistance by uncoupling phosphatidylinositol 3-kinase and its downstream signals in 3T3L1 adipocytes. Diabetes 50:18911900 20. Kopchick JJ, Andry JM 2000 Growth hormone (HG), GH receptor, and signal transduction. Mol Gen Metab 71:293314 21. Dominici FP, Turyn D 2002 Growth homone-induced alterations in the insulin-signaling system. Exp Biol Med 227:149 157 22. Kirwan JP, Hauguel-De Mouzon, Oepercq J, Challier JC, Huston-Presley L, Friedman JE, Kalhan SC, Catalano PM 2002 TNF- is a predictor of insulin resistance in human pregnancy. Diabetes 51:22072213 23. Friedman JE, Ishizuka T, Shao J, Huston L, Highman T, Catalano P 1999 Impaired glucose transport and insulin receptor tyrosine phosphorylation in skeletal muscle from obese women with gestational diabetes. Diabetes 48: 18071814 24. Shao J, Catalano PM, Yamashita H, Ruyter I, Smith S, Youngren J, Friedman JE 2000 Decreased insulin receptor tyrosine kinase activity and plasma cell membrane glycoprotein-1 overexpression in skeletal muscle from obese women with gestational diabetes (GDM). Diabetes 610:603 610 25. Eriksson H, Ridderstrale M, Tornqvist H 1995 Tyrosine phosphorylation of the growth hormone (GH) receptor and Janus tyrosine kinase-2 is involved in the insulin-like actions of GH in primary rat adipocytes. Endocrinology 136:50935101

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

Das könnte Ihnen auch gefallen