Endocrine Journal 2013, 60 (3), 359-368

Or i g i n a l Improvement of insulin sensitivity promotes extravillous trophoblast cell migration stimulated by insulin–like growth factor-I

Reiko Mayama1), Tomoko Izawa1), Keiji Sakai1), Nicolae Suciu2) and Mitutoshi Iwashita1)

1) Department of Obstetrics and Gynecology, Kyorin University School of Medicine, Tokyo 181-8611, Japan 2) Department of Obstetrics and Gynecology, The University of Medicine and Pharmacy “Carol Davila,” The Polizu Hospital, 011062 Bucuresti, Romania

Abstract. Insulin-like growth factor-I (IGF-I) has been shown to stimulate extravillous trophoblast (EVT) cell migration and invasion, and to play a crucial role in placental function, thereby, influencing placental development and fetal growth. Insufficient invasion of EVT cells into the uterine endometrium leads to pregnancy-related complications, including spontaneous abortion, fetal growth restriction (FGR), and pregnancy-induced hypertension (PIH). Insulin-resistant conditions such as polycystic ovary syndrome (PCOS) and gestational diabetes mellitus (GDM) have also been associated with abortion and PIH. However, the effects of IGF-I on EVT cells under insulin-resistant conditions have not been elucidated yet. The current study was undertaken to analyze the effects of IGF-I under insulin-resistant conditions and to determine whether improvement in insulin sensitivity alters IGF signaling and cell migration in the EVT. Incubation with pioglitazone, an insulin sensitizer, increased peroxisome proliferator-activated receptor-γ (PPARγ) expression after 48 h. A 48-h pre- incubation with insulin reduced the phosphorylation and concentration of the insulin receptors, which were increased by insulin treatment. Long-term exposure to insulin reduced phosphorylation of the IGF-I receptor, substrate-1 (IRS-1), and Akt, and also reduced EVT cell migration. However, when the cells were incubated with pioglitazone in addition to insulin for 48 h, the phosphorylation of these was restored. This combination partially reversed the inhibitory effect of insulin on EVT cell migration. These results suggest that abnormalities in pregnancy that are induced by loss of insulin sensitivity can be treated by improving insulin sensitivity.

Key words: IGF-I, Insulin sensitivity, Extravillous trophoblast (EVT), Cell migration

Cytotrophoblast cells differentiate into syncy- Insulin resistance is observed in various conditions tiotrophoblast and extravillous trophoblast (EVT) cells, such as gestational diabetes mellitus (GDM), polycys- and form cell columns during human placental devel- tic ovary syndrome (PCOS), obesity, and increased opment [1]. During early pregnancy, these cell columns weight gain. Several lines of evidence have indicated come into contact with the uterine endometrium, and the that these insulin-resistant conditions are involved in EVT invades the endometrium. Endothelial cells and the pathogenesis of PIH [5-7]. In addition, the inci- medial smooth muscle cells of placental spiral arteries dence of abortion is higher in pregnant women with are invaded by the EVT and remodeled [2]. The remod- PCOS than in those without complications [8, 9]. eling of spiral arteries reduces uteroplacental vascu- Pioglitazone is a thiazolidinedione that acts as a per- lar resistance and increases uteroplacental blood flow. oxisome proliferator-activated receptor-γ (PPARγ) ago- Insufficient invasion of the EVT into the uterine endo- nist. It is essential for glucose metabolism, is involved metrium is reported to lead to pregnancy-related compli- in the insulin signal pathway, and has been used in the cations, including abortion, pregnancy induced hyper- clinical treatment of type 2 diabetes [10]. Pioglitazone tension (PIH) and fetal growth restriction (FGR) [3, 4]. also alters expression, improves insulin sensitiv- ity, and alters glucose intolerance [11, 12]. Submitted Jul. 4, 2012; Accepted Oct. 28, 2012 as EJ12-0241 Insulin-like growth factor-І (IGF-І) stimulates tro- Released online in J-STAGE as advance publication Nov. 30, 2012 Correspondence to: Tomoko Izawa, MD., Ph. D., Department of phoblast cell migration and invasion and plays a cru- Obstetrics and Gynecology, Kyorin University School of Medicine, cial role in placental function, thereby influencing 6-20-6 Shinkawa, Mitaka, Tokyo, 181-8611, Japan placental development and fetal growth [13]. IGF-І E-mail: [email protected] expresses its biological effects through the IGF-I ©The Japan Endocrine Society 360 Mayama et al. receptor (IGF-IR), insulin receptor, and IGF-I/insulin EVT cells were obtained using a previously described hybrid receptor, which are expressed in the placenta method [19], with a few modifications. Briefly, tissue [14-17]. IGF-I phosphorylates tyrosine residues in the was washed with cold phosphate-buffered saline (PBS, beta-subunit of these receptors; these receptors phos- pH 7.4) prior to dissection. The tissue was cut into phorylate insulin receptor substrate-1 (IRS-1) and acti- small pieces and any blood vessels or clots, membranes, vate the mitogen-activated kinase (MAPK) and and decidual tissues were removed. Villous tissue frag- phosphatidyl inositol 3-kinase (PI3K) pathways that ments were cultured with medium 199 supplemented mediate IGF-I actions. Activated PI3K produces PIP3 with 100 μg/mL streptomycin, 100 U/mL penicillin, that directly binds to Akt and mediates cell signaling to 2.5 μg/mL amphotericin B, and 10% fetal calf serum downstream. The PI3K/Akt pathway has been linked (FCS, Sigma-Aldrich Co, St. Louis, MO) on 6-well to several cellular functions, including cell migration cell culture plates, precoated with 20 µg fibronectin [18]. Since the EVT secretes IGF-I and expresses these diluted in 1 mL of PBS. Tissues were allowed to attach receptors, the EVT itself stimulates migration and pro- to the bottom of the plate for 2 h before adding 200 µL liferation in an autocrine manner [13]. However, lit- of the same medium. One milliliter of growth medium tle is known about the effect of insulin-resistant condi- 199 supplemented with 100 μg/mL streptomycin, 100 tions on IGF-I activity in the EVT. U/mL penicillin, 2.5 μg/mL amphotericin B, and 10% Therefore, we analyzed IGF-I activity in terms of FCS was added after 24 h, and culture was continued IGF-I signaling and cell migration under insulin-resis- for an additional 3-5 days. EVT cells were spread out tant condition with the EVT and further evaluated the around tissues during culture, original placental tissues improvement of insulin sensitivity using pioglitazone were then removed, and culture was continued for an could alter the IGF-I signaling, and resulted in promot- additional 1-2 weeks. The medium was changed every ing EVT migration. 48 h until cells were confluent in 5% CO2/95% air at 37 °C. Cells were passaged using trypsin. EVT cells Materials and methods between passages 2-4 were used for all experiments. The cells were identified as EVT cells by- immu Materials nohistochemical staining using anti-cytokeratin 7 and Tissue culture media, penicillin, streptomycin, 8/18, anti-α5β1 and αvβ3 , anti-vimentin, and amphotericin B were purchased from Sigma- CD9, and factor VIII [20, 21]. More than 90% cells Aldrich Co. (St. Louis, MO). Polyvinylidenedifluoride expressed all these EVT markers. (PVDF) membranes (Immobilon-P) were purchased from Millipore (Bedford, MA). The following anti- Detection of PPARγ expression bodies were purchased from Santa Cruz Biotechnology Some studies have shown that PPARγ promotes Co. (Santa Cruz, CA): anti-phosphotyrosine (PY99), β the expression of several proteins, including PPARγ subunit of IGF-I receptor (β-IGF-IR) (C-20), β subunit itself [22, 23]. To determine the activation of PPARγ, of insulin receptor (IR-β) (C-19), anti-insulin receptor expression of PPARγ was measured in the EVT. The substrate-1 (IRS-1) (A-19), anti-phophoAkt-1/2/3, and cells were incubated with 10 μM pioglitazonefor 0-48 anti-Akt-1. Anti-PPARγ antibody was purchased from h. Then, nuclear extracts were prepared and immuno- Aviva Systems Biology (San Diego, CA). Recombinant blotting was performed. For nuclear extraction, EVTs human IGF-I was a gift from Fujisawa Pharmaceutical were incubated on ice for 10 min with 2 mL of buffer A (Osaka, Japan). All other chemicals were purchased (10 mM HEPES-KOH, 1.5 mM MgCl2, 10 mM KCl, 1 from Sigma-Aldrich Co. (St. Louis, MO). mM DTT, and the protease inhibitor mixture Complete, pH 7.9 [Sigma-Aldrich Co.]). The cells were then col- Primary culture of EVT cells lected, homogenized, and centrifuged at 2,500 × g for Normal first trimester placenta (6-10 weeks) was 5 min at 4 °C. The supernatant was removed and the obtained in cases of legal elective termination of pellet was homogenized with 1 mL of buffer B (0.25 pregnancy. All the patients provided informed con- M sucrose, 1% TritonX-100, 10 mM Tris-HCl, 1.5 mM sent for collection and investigational use of tissues. MgCl2, 1 mM DTT and protease inhibitor mixture, pH This study was approved by the ethics committee of 7.4). The same volume of buffer C (0.5 M sucrose, 1% Kyorin University, School of Medicine, Tokyo, Japan. TritonX-100, 10 mM Tris-HCl, 1.5 mM MgCl2, 1 mM Insulin sensitivity alters IGF-I effects 361

DTT, and protease inhibitor mixture, pH 7.4) was then to 50% confluence. The cells were incubated for 48 h added and centrifuged at 3,000 × g for 10 min at 4 °C. in medium 199 containing 3.0 g/L of glucose, 2% FCS, The supernatant was removed and the pellet was resus- antibiotics (100 μg/mL streptomycin, 100 U/mL peni- pended in 25 μL of sample buffer. cillin, and 2.5 μg/mL amphotericin B), with or without 30 nM insulin in the absence or the presence of 10 μM Reduction of insulin sensitivity pioglitazone. The confluent monolayers were wounded To achieve insulin-resistant conditions, insulin pre- with a single-edge razor blade, leaving a denuded area incubation was performed [24]. EVT cells were incu- and a sharp visible demarcation line at the wound edge, bated with or without 1, 10, and 30 nM insulin for 48 h as described previously [21]. The wounded monolay- in low-serum medium (3.0 g/L glucose and 2% FCS). ers were rinsed twice with serum-free medium (SFM) After pre-incubation, the presence of insulin-resistant 199 containing 3.0 g/L glucose. The SFM was then conditions was confirmed by phosphorylation of IRβ replaced with SFM with or without 100 ng/mL IGF-I, after stimulation with 10 nM insulin for 5 min. The in the absence or presence of 10 μM pioglitazone. The cells were then used for immunoprecipitation and a cells were incubated for an additional 48 h. At the end migration assay. of incubation, the cells were rinsed in PBS, fixed and stained using the Diff-Quick (International Reagents Immunoprecipitation and immunoblotting Co., Kobe, Japan), and examined using phase-contrast The methods used for immunoprecipitation have microscopy. The migrated EVTs were counted in 1.0- been described previously [20, 21]. EVT cells were mm-long sections, allowing for a 20-μm space from the washed with cold PBS and lysed in lysis buffer (50 mM demarcation line in order to minimize the possible phys- Tris-HCl [pH 7.5], 150 mM NaCl, 1% Nonidet P-40, 2 ical effects of cell movement resulting from cell prolif- mM EGTA, 100 mM sodium fluoride, 10 mM sodium eration. A calibrated eyepiece grid was used to deter- pyrophosphate, 2 mM sodium vanadate, 1 mM 4-(2- mine the mean number of migrated cells. The results aminoethyl) benzenesulfonyl fluoride [AEBSF], 1 μg/ were calculated by averaging a mean of 10 sections mL pepstatin A, 1 μg/mL leupeptin, 1 μg/mL aprotinin). per test substance for each experiment. The number of The cell lysates were centrifuged at 15,000 × g for 10 migrated cells was expressed in terms of mean ± SD. min at 4 °C. The supernatants were incubated with the corresponding antibodies overnight at 4 °C. The immu- Statistical analysis nocomplexes were incubated with 35 μL of Protein-A Statistical analysis was performed using Stat View Sepharose for 2 h. The immunoprecipitates were 5.0 (SAS Institute Inc., Cary, NC) and one-way washed 3 times with lysis buffer. The bound proteins ANOVA followed by post hoc test was used to com- were eluted in 25 μL of Laemmli sample buffer, heated pare the differences amongtest groups (p < 0.05 was for 10 min at 95 °C, and then analyzed by SDS-PAGE considered statistically significant). The experiments on 7.5% acrylamide gels. The separated proteins were were repeated at least 3 times in each group to assess electrically transferred to a PVDF membrane (pore size, reproducibility. All values represented the mean ± S.D. 0.45 μm). The membranes were incubated overnight of three separated experiments. at 4 °C with the antibodies indicated in the figures and figure legends. The immunocomplexes were detected Results using either an anti-mouse or anti-rabbit horseradish peroxidase-conjugated antibody and visualized using Effect of pioglitazone on PPARγ expression enhanced chemiluminescence, following the manu- EVT cells were incubated with 10 μM pioglitazone- facturer’s instructions (Super-Signal CL-H substrate for 12, 24, and 48 h, and PPARγ expression was mea- system; Pierce, Rockford, IL). The images obtained sured. PPARγ expression increased to 1.1, 3.0, and 4.8 were scanned using EZ capture II (ATTO Corporation, fold, respectively, of that at 0 h (Fig. 1). Since 48 h-in- Tokyo, Japan). Densitometric analyses of the images cubation with pioglitazone increased PPARγ expres- were performed using ImageJ (version 1.41). sion significantly (p<0.01, compared with control), in subsequent experiments, EVTs were incubated with Cell migration assay pioglitazone for 48 h to improve insulin sensitivity. EVT cells were plated on 6-well plates and grown The scanning densitometry values of the bands were 362 Mayama et al.

Fig. 1 Effect of pioglitazone on PPARγ expression EVT cells were grown to 80% confluency. Then, the cells were incubated with pioglitazone (10μM) for 0-48 h. After the incubation, nuclear extracts were prepared as described under “Material and Methods”. The nuclear extracts were analyzed by SDS-PAGE followed by immunoblotting using anti PPARγ antibody.

Fig. 2 Effect of insulin pre-incubation on insulin receptor phosphorylation and insulin receptor expression EVT cells were incubated with various concentrations of insulin (0-30 nM) for 48 h and stimulated with 10nM insulin for 5 min. Then cells were lysed and immunoprecipitated with anti-insulin receptor β (IRβ) subunit antibody. The proteins were analyzed by SDS-PAGE followed by immunoblotting using PY-99 (upper panel) and anti IRβ subunit antibodies (lower panel).

6320.7±993.2, 6804.7±1069.2, 19233.7±3022.2 and insulin for 48 h to obtain insulin-resistant conditions. 30379.9±4773.7, from left to right respectively. Effect of pioglitazone on the insulin receptor Effect of insulin pre-incubation on insulin receptor To evaluate the effect of pioglitazone on the insulin- phosphorylation and insulin receptor expression resistant condition of EVTs, pioglitazone (10 μM) was Pre-incubation of EVT cells with 1 nM insulin had coincubated with insulin (30 nM) for 48 h to investi- no effect; however, pre-incubation with 10 nM and 30 gate phosphorylation of the insulin receptor following nM insulin significantly reduced phosphorylation of IRβ stimulation with 10 nM insulin for 5 min. Stimulation stimulated by 10 nM insulin to 57.7% (p<0.01, com- with 10 nM insulin for 5 min increased phosphoryla- pared with control) and 47.8% (p<0.01, compared with tion of the insulin receptor but in the cells that were control) , respectively (Fig. 2 upper panel, lanes 3 and pre-incubated with 30 nM insulin for 48 h, phosphory- 4). The scanning densitometry values of the upper panel lation of the insulin receptor was suppressed (26449.2± were 11888.8±2475.6, 10284.6±1442.6, 6863.4±887.2 2004.5 vs. 18627.4±152.9 [Fig. 3, upper panel, lane2 and 5693.9±737.7, from left to right, respectively. Pre- vs. lane 4]; p < 0.001). When 10 μM pioglitazone was incubation with 10 and 30 nM insulin also reduced the coincubated with 30 nM insulin for 48 h, insulin recep- total concentrations of IRβ to 44.8% (p<0.05, compared tor phosphorylation stimulated by 10 nM insulin for 5 with control) and 43.9% (p<0.05, compared with con- min was restored (18627.4±152.9 vs. 21350.9±1181.8 trol), respectively (Fig. 2 lower panel, lanes 3 and 4). The [Fig. 3, upper panel, lane 4 vs. lane 6]; p < 0.05). scanning densitometry values of the lower panel were The concentration of the insulin receptor was also 27384.5±949.0, 247820.2±4640.1, 12274.9±1319.3 suppressed by insulin pre-incubation (Fig. 3, lower and 12036.3.9±1963.8, from left to right. These results panel, lanes 3 and 4) compared to that in the control were suggesting that insulin pre-incubation consid- cells (Fig. 3, lower panel, lanes 1 and 2). However, erably elicits insulin-resistant conditions. For further when pioglitazone was coincubated with 30 nM insu- experiments, EVT cells were pre-incubated with 30 nM lin, the insulin receptor concentration was restored Insulin sensitivity alters IGF-I effects 363

Fig. 3 Effect of pioglitazone on the insulin receptor EVT cells were incubated with 30 nM insulin (lane 3-6) and 10μM pioglitazone (lane 5, 6) for 48 h and stimulated with 10nM insulin for 5 min. Then cells were lysed and immunoprecipitated with anti IR antibody. The proteins were analyzed by SDS- PAGE followed by immunoblotting using PY-99 (upper panel) and anti IR antibodies (lower panel). *p<0.05, **p<0.01

(Fig. 3, lower panel, lanes 5 and 6), suggesting that pio- panel, lane 2) that was reduced in cells treated with glitazone improved the insulin sensitivity of EVTs. 30 nM insulin before incubation (26835.0±520.3 vs. 12227.4±1207.3 [Fig. 5, upper panel, lane 2 vs. lane Effects of insulin sensitivity on signal transduction by 4]; p < 0.001). However, if EVT cells were coincu- IGF-I bated with 10 μM pioglitazone and 30 nM insulin, To investigate the effects of insulin sensitivity on IRS-1 phosphorylation stimulated by 100 ng/mL IGF-I IGF-I signaling, phosphorylation of IGF-IR, IRS-1, improved (12227.4±1207.3 vs. 17208.1±1455.8 [Fig. and Akt was examined after preincubation with insu- 5, upper panel,lane4 vs. lane 6], p < 0.01). The total lin and pioglitazone for 48 h. IGF-I at 100 ng/mL con- concentrations of IRS-1 were similar in all lanes (Fig. centration increased the phosphorylation of IGF-IR 5, lower panel). as observed in the control cells (Fig. 4, upper panel, IGF-I promoted the phosphorylation of Akt (Fig. 6, lane 2). In the cells with insulin preincubation, IGF-IR upper panel, lane 2), and insulin pre-incubation also phosphorylation stimulated by IGF-I (100 ng/mL) was inhibited Akt phosphorylation stimulated by 100 ng/ lower than that in the control cells (15677.7±2131.7 vs. mL IGF-I (2203807±1870.4 vs. 9507.2±1446.5 [Fig. 6, 11427.2±1966.6 [Fig. 4, upper panel, lane 2 vs. lane 4], upper panel, lane 2 vs. lane 4] p < 0.01). However, pre- p < 0.05). When 10 μM pioglitazone was coincubated incubation with 10 μM pioglitazone and 30 nM insulin with 30 nM insulin, IGF-IR phosphorylation stimu- improved this inhibitory effect on Akt phosphorylation lated by IGF-I was increased to a level greater than (9507.2±1446.5 vs. 15631.8±537.7 [Fig. 6, upper panel, that in the cells without pioglitazone (11427.2±1966.6 lane4 vs. lane 6] p < 0.05), whereas Akt concentrations vs. 19112.3±2513.5 [Fig. 4, upper panel,lane4 vs. lane were similar in all the lanes (Fig. 6, lower panel). 6], p < 0.01) . However, total concentrations of IGF-IR were similar in all the lanes (Fig. 4, lower panel). Effect of insulin pre-incubation on EVT cell migra- Similarly, stimulation with 100 ng/mL IGF-I also tion stimulated by IGF-I increased the phosphorylation of IRS-1 (Fig. 5, upper IGF-I (100 ng/mL) substantially stimulated migra- 364 Mayama et al.

Fig. 4 Effects of insulin prei-ncubation on IGF-IR phosphorylation EVT cells were incubated with 30 nM insulin (lane 3-6) and 10μM pioglitazone (lane 5, 6) for 48 h and stimulated with IGF-I (100 ng/mL) for 10 min. Then cells were lysed and immunoprecipitated with anti IGF-I antibody. The proteins were analyzed by SDS-PAGE followed by immunoblotting using PY-99 (upper panel) and anti IGF-I antibodies (lower panel). *p<0.05

Fig. 5 Effects of insulin pre-incubation on IRS-1 phosphorylation EVT cells were incubated with insulin (30 nM) (lane 3-6) and pioglitazone (10 μM) (lane 5, 6) for 48 h and stimulated by 100 ng/mL IGF-I for 10 min. Then cells were lysed and immunoprecipitated with anti-IRS-1antibody. The proteins were analyzed by SDS-PAGE followed by immunoblotting using PY-99 (upper panel) and anti IRS-1antibodies (lower panel). *p<0.01, **p<0.001 Insulin sensitivity alters IGF-I effects 365

Fig. 6 Effects of insulin pre-incubation on Akt phosphorylation EVT cells were incubated with 30 nM insulin (lane 3-6) and 10 μM pioglitazone (lane 5, 6) and incubated for 48 h and cells were stimulated by 100 ng/mL IGF-I for 10 min. Then the cells were lysed and the proteins were analyzed by SDS-PAGE followed by immunoblotting using anti-phosphoAkt antibody (upper panel) and anti Akt-1 antibodies (lower panel). *p<0.05, **p<0.01

Fig. 7 Effect of insulin pre-incubation on EVT cell migration stimulated by IGF-I EVT cells were incubated with medium 199 containing 3.0g/L glucose, 10% FCS, 30nM of insulin and 10μM pioglitazone. The confluent cell monolayers were wounded with a single edge razor blade. Then the medium was replaced with SFM, containing 100ng/mL IGF-I (lane2, 4, 6) and 10μM pioglitazone (lane 5, 6). The cells were incubated for additional 48 h, and then the migrated cells were counted. The results are expressed as ratio to control. 366 Mayama et al. tion of EVT cells compared to that for the control more pivotal role in IGF-I signaling than IGF-IR in the (Fig. 7); cell migration of EVT cells was more than EVT [15]. Therefore, it is possible that suppression of 2-fold higher than that for the control (1.00±0.19 vs. IGF-I action, due to insulin-resistant conditions, is due 2.28±0.23, p< 0.001). In contrast, insulin pre-incuba- to decrease in the IGF-I/insulin hybrid receptor, which tion significantly inhibited IGF-I stimulated cell migra- mediates IGF-I signaling. Another possible explana- tion (1.35±0.18-fold lower than the control; p < 0.01). tion for reduced IGF-I action in insulin-resistant con- Pre-incubation with insulin (30 nM) and pioglitazone ditions is that insulin pre-incubation directly decreases (10 μM) restored the inhibitory effect of insulin pre-in- phosphorylation of tyrosine residues in IRS-1. Hemi cubation (1.36±0.18 vs. 1.69±0.18, p < 0.05). et al. have shown that cells incubated with insulin for a long period enhance phosphorylation of serine resi- Discussion dues in IRS-1, which reduces the phosphorylation of tyrosine residues in IRS-1 by insulin [30]. Insulin pre-incubation decreases both insulin recep- Pioglitazone, a PPARγ agonist, is used in the treat- tor concentrations and insulin-stimulated tyrosine phos- ment of type 2 diabetes [31]. It alters gene transcrip- phorylation of the insulin receptor (Fig. 2), indicating tion related to glucose and lipid metabolism, which that insulin pre-incubation resulted in insulin-resistant is related to the production of several proteins such conditions in the EVT. Pre-incubation with insulin for as glucose transporters, resulting in improved glu- 48 h stimulates internalization of insulin receptors on the cose uptake [32]. It has been reported that pioglita- cell surface, thereby desensitizing EVT against further zone potentiates the activation of PPARγ and increases insulin stimulation. In general, attenuation of insulin the expression of PPARγ itself [23]. Seto-Young et al. sensitivity is caused not only by decreasing the receptor have shown that pioglitazone stimulates PPARγ expres- number on the cell surface, but also by various mecha- sion and increases IR expression in ovarian cell culture nisms, including malfunction of receptors and impaired [22]. Therefore, it is reasonable to assume that PPARγ signal pathways [25, 26]. High insulin levels in the increases the expression of the IR in EVT cells, which circulation are frequently associated with GDM [27]. may increase IGF-I/insulin hybrid receptor expression, Therefore insulin-receptor deprivation due to excess thereby improving IGF-I signaling. Pioglitazone is insulin may one of the main pathogenic mechanisms by prohibited to prescribe for pregnant woman. Therefore which insulin resistance is elicited in GDM. In addi- it is not practical to use pioglitazone for improve- tion, several studies have used insulin pre-incubation ment of insulin sensitivity during pregnancy. But for a to establish insulin-resistant conditions in their exper- woman who has insulin resistance and plans to be preg- iments [24], thus we also used insulin pre-incubation nant, pioglitazone is useful for improvement of insulin to achieve insulin-resistant conditions, although down- sensitivity. And administration of pioglitazone might regulation of insulin receptor has not been reported in reduce pregnancy related complications in the future. pregnant woman with hyperinsulinemia. In this study, IGF-I was found to stimulate migration The current study shows that insulin pre-incuba- of EVT cells. We have previously reported the mecha- tion suppresses tyrosine phosphorylation of IGF-IR, nism by which IGF-I stimulated EVT migration. Both resulting in the suppression of IRS-1 and Akt phospho- αvβ3- [20] and α5β1-integrin [21] are involved rylation stimulated by IGF-I. IGF-IR and the insulin in EVT cell migration, and IGF-I acts by activating the receptor, both of which are receptors, integrin signal pathways in EVT cells. Since the IGF-I/ share 60% homology, and these receptors can het- insulin hybrid receptors are abundant on EVT cell sur- erodimerize into IGF-I/insulin hybrid receptors com- face, and play a more pivotal role than IGF-IR for posed of one IGF-IR αβ-dimer and one IR αβ-dimer. IGF-I signaling in the EVT [15], hybrid receptor-me- The IGF-I/insulin hybrid receptor has high affinity diated cell signaling may cross talk with integrin sig- for the IGF-I, but not insulin [14, 28]. The EVT pos- naling, which may be one of the reasons why EVT cell sesses considerable amounts of IGF-I/insulin hybrid migration by IGF-I is inhibited under insulin-resistant receptors [29] as well as the IR and IGF-IR. Insulin conditions. preincubation decreases the concentration of the IR In the current study, pioglitazone restored EVT cell αβ-dimer, which may lead to a decrease in the IGF-I/ migration stimulated by IGF-I. This is consistent with insulin hybrid receptor. The hybrid receptors play a the result that pioglitazone restores inhibition of IGF-I Insulin sensitivity alters IGF-I effects 367 signaling induced by insulin pre-incubation. EVT cell In conclusion, the current studies have shown that migration is an important event for the establishment the effects of IGF-I signaling are attenuated in insu- and maintenance of pregnancy. One of the insulin sen- lin-resistant conditions in EVT cells, and improvement sitizers, metformin, has been reported to increase preg- in insulin sensitivity by pioglitazone alters IGF signal- nant rates after in vitro fertilization [33] and reduce ing, resulting in the promotion of EVT cell migration. miscarriage rates in patients with polycystic ovary syn- Improvement in insulin sensitivity contributes to nor- drome [34]. Thus, improvement in insulin resistance malization of placental development and may reduce could also improve the clinical outcomes of pregnancy the incidence of abortion, PIH, and FGR in insulin-re- under insulin-resistant conditions. sistant conditions.

References

1. Bischof P, Irminger-Finger I (2005) The human cytotro- its treatment by thiazolidinediones. Recent Prog Horm phoblastic cell, a mononuclear chameleon. Int J Biochem Res 56:265-294. Cell Biol 37:1-16. 12. Iwata M, Haruta T, Usui I, Takata Y, Takano A, Uno T, 2. Lockwood CJ, Huang SJ, Krikun G, Caze R, Rahman Kawahara J, Ueno E, Sasaoka T, Ishibashi O, Kobayashi M, Buchwalder LF, Schatz F (2011) Decidual hemosta- M (2001) Pioglitazone ameliorates tumor necrosis fac- sis, inflammation, and angiogenesis in pre-eclampsia. tor-alpha-induced insulin resistance by a mechanism Semin Thromb Hemost 37:158-164. independent of adipogenic activity of peroxisome pro- 3. Kaufmann P, Black S, Huppertz B (2003) Endovascular liferator--activated receptor-gamma. Diabetes 50:1083- trophoblast invasion: implications for the pathogene- 1092. sis of intrauterine growth retardation and preeclampsia. 13. Lacey H, Haigh T, Westwood M, Aplin JD (2002) Biol Reprod 69:1-7. Mesenchymally-derived insulin-like growth factor 1 4. Meekins JW, Pijnenborg R, Hanssens M, McFadyen IR, provides a paracrine stimulus for trophoblast migration. van Asshe A (1994) A study of placentalbedspiralarter- BMC Dev Biol 24:2-5. ies and trophoblastinvasion in normal and severepre- 14. Bäck K, Brännmark C, Strålfors P, Arnqvist HJ (2011) eclampticpregnancies. Br J Obstet Gynaecol 101:669- Differential effects of IGF-I, IGF-II and insulin in human 674. preadipocytes and adipocytes--role of insulin and IGF-I 5. Solomon CG, Seely EW (2001)Hypertension in preg- receptors. Mol Cell Endocrinol 339:130-135. nancy: A manifestation of insulin resistance syndrome? 15. Hashimoto R, Sakai K, Matsumoto H, Iwashita M Hypertension 37:232-239. (2010) Tumor necrosis factor-alpha (TNF-alpha) inhib- 6. Miyakoshi K, Tanaka M, Matsumoto T, Hattori Y, Ueno its insulin-like growth factor-I (IGF-I) activities in K, Teranishi T, Minegishi K, Ishimoto H, Shimada A, human trophoblast cell cultures through IGF-I/insulin Yoshimura Y(2004) Hypertensive disorders in Japanese hybrid receptors. Endocr J 57:193-200. women with gestational glucose intolerance. Diabetes 16. Johansson GS, Arnqvist HJ (2006) Insulin and IGF-I Res Clin Pract 64:201-205. action on insulin receptors, IGF-I receptors, and hybrid 7. Hidaka A, Hukuda H, Imoto H, Yamazaki T, Nishimura insulin/IGF-I receptors in vascular smooth muscle cells. J, Nakamoto O (2008) Risk factors for preeclampsia. Am J Physiol Endocrinol Metab 291: E1124-1130. Obstetrical and Gynecological Therapy 103:646-647 17. VandanaSaini (2010) Molecular mechanisms of insulin (In Japanese). resistance in type 2 diabetes mellitus. World J Diabetes 8. Sohrabvand F, Shariat M, Haghollahi F, Bagheri B 15:68-75. (2009) Effect of metformin on miscarriage in pregnant 18. Clemmons DR, Maile LA (2003) Minireview: Integral patients with polycystic ovary syndrome. West Indian membrane proteins that function coordinately with the Med J 58: 433-436. insulin-like growth factor I receptor to regulate intracel- 9. Chen ZJ, Shi Y (2010) Polycystic Ovary Syndrome. lular signaling. Endocrinology 144:1664-1670. Front Med China 4: 280-284. 19. Yagel S, Casper RF, Powell W, Parhar RS, Lala PK 10. Higashi Y, Holder K, Delafontaine P (2010) (1989) Characterization of pure human first-trimester Thiazolidinediones up-regulate insulin-like growth fac- cytotrophoblast cells in long-term culture: growth pat- tor-1 receptor via a peroxisome proliferator-activated tern, markers, and hormone production. Am J Obstet receptor gamma-independent pathway. J Biol Chem Gynecol 160: 938-945. 285: 36361-36368. 20. Kabir-Salmani M, Shiokawa S, Akimoto Y, Sakai K, 11. Lebovitz HE, Banerji MA (2001) Insulin resistance and Nagamatsu S, Sakai K, Nakamura Y, Lotfi A, Kawakami 368 Mayama et al.

H, Iwashita M (2003) Alphavbeta3 integrin signal- tational diabetes. Adv Med Sci 56: 200-206. ing pathway is involved in insulin-like growth factor 28. Sakai K, Lowman HB, Clemmons DR (2002) Increases I-stimulated human extravilloustrophoblast cell migra- in free, unboundinsulin-like growth factor I enhance tion. Endocrinology 144:1620-1630. insulin responsiveness in human hepatoma G2 cells in 21. Kabir-Salmani M, Shiokawa S, Akimoto Y, Sakai K, culture. J Biol Chem 277:13620-13627. Iwashita M (2004) The role of alpha(5)beta(1)-integrin 29. Treadway JL, Morrison BD, Goldfine ID, Pessin JE in the IGF-I-induced migration of extravilloustropho- (1989) Assembly of insulin/insulin-like growth fac- blast cells during the process of implantation. Mol Hum tor-1 hybrid receptors in vitro. J Biol Chem 264: 21450- Reprod 10:91-97. 21453. 22. Seto-Young D, Avtanski D, Strizhevsky M, Parikh G, 30. Hemi R, Yochananov Y, Barhod E, Kasher-Meron M, Patel P, Kaplun J, Holcomb K, Rosenwaks Z, Poretsky Karasik A, Tirosh A, Kanety H (2011) p38 mitogen- L(2007) Interactions among peroxisome proliferator activated -dependent transactivation of activated receptor-gamma, insulin signaling pathways, ErbBreceptor family: a novel common mechanism for and steroidogenic acute regulatory protein in human stress-induced IRS-1 serine phosphorylation and insulin ovarian cells. J Clin Endocrinol Metab 92: 2232-2239. resistance. Diabetes 60:1134-1145. 23. Panchapakesan U, Pollock CA, Chen XM (2004) The 31. DeFronzo RA, Tripathy D, Schwenke DC, Banerji M, effect of high glucose and PPAR-gamma agonists on Bray GA, Buchanan TA, Clement SC, Henry RR, Hodis PPAR-gamma expression and function in HK-2 cells. HN, Kitabchi AE, Mack WJ, Mudaliar S, Ratner RE, Am J Physiol Renal Physiol 287: F528-534. Williams K, Stentz FB, Musi N, Reaven PD (2011) 24. Ronnett GV, Knutson VP, Lane MD (1982) Insulin- Pioglitazone for diabetes prevention in impaired glu- induced down-regulation of insulin receptors in 3T3-L1 cose tolerance. N Engl J Med 364:1104-1115. adipocytes. Altered rate of receptor inactivation. J Biol 32. Koenen TB, Tack CJ, Kroese JM, Hermus AR, Sweep Chem 257:4285-4291. FC, van der Laak J, Stalenhoef AF, de Graaf J, van Tits 25. Young J, Morbois-Trabut L, Couzinet B, Lascols LJ, Stienstra R (2009) Pioglitazone treatment enlarges O, Dion E, Béréziat V, Fève B, Richard I, Capeau J, subcutaneous adipocytes in insulin-resistant patients. J Chanson P, Vigouroux C (2005) Type A insulin resis- Clin Endocrinol Metab 94: 4453-4457. tance syndrome revealing a novel lamin A mutation. 33. Jinno M, Kondou K, Teruya K (2010) Low-dose met- Diabetes 54: 1873-1878. formin improves pregnancy rate in in vitro fertilization 26. Wang RH, Kim HS, Xiao C, Xu X, Gavrilova O, Deng repeaters without polycystic ovary syndrome: predic- CX (2011) Hepatic Sirt1 deficiency in mice impairs tion of effectiveness by multiple parameters related to mTorc2/Akt signaling and results in hyperglycemia, insulin resistance. Hormones (Athens) 9:161-170. oxidative damage, and insulin resistance. J Clin Invest 34. Sohrabvand F, Shariat M, Haghollahi F, Bagheri B 121: 4477-4790. (2009) Effect of metformin on miscarriage in pregnant 27. Matuszek B, Lenart-Lipińska M, Burska A, Paszkowski patients with polycystic ovary syndrome. West Indian T, Smoleń A, Nowakowski A (2011) Increased serum Med J 58: 433-436. insulin-like growth factor-1 levels in women with ges-