Objective
Diabetes mellitus is a risk factor for preeclampsia. Cytotrophoblast (CTB) invasion is facilitated from the conversion of plasminogen to plasmin by urokinase plasminogen activator (uPA), regulated by plasminogen activator inhibitor 1 (PAI-1), and may be inhibited in preeclampsia. This study assessed signaling mechanisms of hyperglycemia-induced CTB dysfunction.
Study Design
Human CTBs were treated with 45, 135, 225, 495, or 945 mg/dL glucose for 48 hours. Some cells were pretreated with a p38 inhibitor (SB203580) or a peroxisome proliferator-activated receptor-gamma (PPAR-γ) ligand (rosiglitazone). Expression of uPA, PAI-1, and PPAR-γ levels and p38 mitogen-activated protein kinase phosphorylation were measured by Western blot in cell lysates. Messenger ribonucleic acid of uPA and PAI-1 was measured by quantitative polymerase chain reaction. Levels of interleukin-6, angiogenic (vascular endothelial growth factor [VEGF], placenta growth factor [PlGF]) and antiangiogenic factors (soluble fms-like tyrosine kinase-1 [sFlt-1], soluble endoglin [sEng]) were measured in the media by enzyme-linked immunosorbent assay kits. Statistical comparisons were performed using analysis of variance with a Duncan’s post-hoc test.
Results
Both uPA and PAI-1 protein and messenger ribonucleic acid were down-regulated ( P < .05) in CTBs treated with 135 mg/dL glucose or greater compared with basal (45 mg/dL). The sEng, sFlt-1, and interleukin-6 were up-regulated, whereas the VEGF and PlGF were down-regulated by 135 mg/dL glucose or greater. p38 phosphorylation and PPAR-γ were up-regulated ( P < .05) in hyperglycemia-treated CTBs. The SB203580 or rosiglitazone pretreatment showed an attenuation of glucose-induced down-regulation of uPA and PAI-1.
Conclusion
Hyperglycemia disrupts the invasive profile of CTB by decreasing uPA and PAI-1 expression; down-regulating VEGF and PlGF; and up-regulating sEng, sFlt-1, and interleukin-6. Attenuation of CTB dysfunction by SB203580 or rosiglitazone pretreatment suggests the involvement of stress signaling.
Preeclampsia (preE) is a complex syndrome that is produced by various pathophysiological triggers and mechanisms affecting 3-8% of pregnancies worldwide. PreE has a higher incidence in women with diabetes mellitus than in the nondiabetic population (about 1 in 5 vs about 1 in 20, respectively). Although the pathogenesis of preE is largely unknown, many investigations have focused on the incompleteness of placental invasion of cytotrophoblast (CTB) cells ; these same explorations have yet to be studied with reference to diabetes.
CTB cells are essential to the development of a successful pregnancy. During the first trimester, CTB cells take on an invasive character as extravillous CTB cells invade and reduce resistance within the spiral arteries, thus allowing for optimal maternal blood transport to the placenta. Disrupted invasion by CTB cells hinders arterial remodeling, causing shallow placentation, which predisposes to preeclampsia.
Extracellular matrix digestion via proteinase activation of the plasmin pathway facilitates CTB invasiveness in the endometrium. Urokinase plasminogen activator (uPA) acts independently of fibrin and is involved in the regulation of cell adhesion and migration of trophoblastic cells. The uPA messenger ribonucleic acid (mRNA) and immunoreactivity have been detected in rhesus monkey CTB cells and in first- and third-trimester human decidual cells. The expression of plasminogen activator inhibitor-1 (PAI-1) in CTB cells is integral in establishing a functional maternal-fetal interface and has been demonstrated to be affected by the p38 pathway.
CTB proliferation, differentiation, invasiveness, and apoptosis are all influenced directly by mitogen-activated protein kinase (MAPK) and indirectly by interleukin-6 (IL-6) during times of cellular stress. Peroxisome proliferator-activated receptor gamma (PPAR-γ), a subtype of the ligand-activated transcription factor superfamily, stimulates villous trophoblastic differentiation and proliferation and has been evaluated as a potential therapeutic target in preeclampsia. In normal pregnant serum, PPAR-γ activators up-regulate the expression and activity of the PPAR-γ receptor. Inversely, PPAR-γ agonists, such as rosiglitazone, cause inhibition of extravillous CTB cell invasion through competitive binding with the retinoid X receptor-α heterodimers in vitro.
Revascularization at the placental interface also involves multiple regulatory pathways of angiogenic and antiangiogenic factors. Vascular endothelial growth factor (VEGF) promotes the syncytialization and proliferation of extravillious trophoblast. Placental growth factor (PlGF), a member of the VEGF subfamily, is expressed by CTB cells and is fundamental for angiogenesis. Antiangiogenic markers include soluble fms-like tyrosine kinase-1 (sFlt-1), a VEGF receptor antagonist), and the capillary tube inhibitor soluble endoglin (sEng).
In previous studies, our research team demonstrated a correlation between levels of a marinobufagenin, a urinary marker elevated in patients with preeclampsia, and an angiogenic imbalance of the markers mentioned in previous text in relation to cardiotonic steroids. Consistent with other studies, this demonstrated that marinobufagenin inhibits CTB proliferation, migration, and invasion. Using similar methodology, we evaluated how hyperglycemia-induced stress signaling has an impact on the invasive phenotype and angiogenic balance of CTB cells to help better understand the relationship between diabetes and preeclampsia.
Materials and Methods
CTB cell culture
The human extravillous CTB cell line Sw.71 utilized in these studies was derived from first-trimester chorionic villus tissue and was kindly provided by Dr Gil G. Mor (Yale University School of Medicine, New Haven, CT). These cells are well characterized and share many characteristics with isolated primary cells, including the expression of cytokeratin-7, human leukocyte antigen class I antigen, human leukocyte antigen-G, BC-1, CD9, human chorionic gonadotropin, and human placental lactogen. Sw.71 cells were cultured in Dulbecco’s modified Eagle’s medium/F-12 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 10 mM N-2-hydroxyethylpipera-zine-N’-2-ethane sulfonic acid, 0.1 mM minimal essential medium nonessential amino acids, 1 mM sodium pyruvate, and 100 U/mL penicillin/streptomycin. Cells were incubated at 37°C, 6% CO 2 , 5% O 2 , and 99% humidity (Isotemp CO 2 incubator; Fisher, Waltham, MA), with no exposure to hypoxic conditions.
Effect of hyperglycemia on CTB cells
CTB cells were seeded on 6-well plates. Prior to treatment, cells were incubated in serum-free media for 24 hours. Cells were treated with 45 (basal), 135, 225, 495, or 945 mg/dL of glucose (Sigma, St. Louis, MO) for 72 hours. Each plate had 2 wells assigned to the basal condition and 1 well assigned to each of the 4 increased glucose levels. Thus, 1 plate constituted 1 replicate for a series of glucose exposures. Some cell suspensions were pretreated with 10 μM of a p38 inhibitor (SB203580) or 10 μM of a PPAR-γ ligand (rosiglitazone) for 3 hours prior to seeding into plates with the glucose treatments.
Antibodies and primers
Primary antibodies
The following antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA): uPA (H-140), catalog no. sc-14019, concentration, 1:500; PAI-1 (H-135), catalog no. sc-8979, concentration, 1:500; PPAR-γ (H-100), catalog no. sc-7196, concentration, 1:500; p-p38 (Tyr 182)-R, catalog no. sc-7975-R, concentration, 1:500; p38α/β (H-147), catalog no. sc-7149, concentration, 1:500; and β-actin (C4), catalog no. sc-47778, concentration, 1:10,000.
Secondary antibodies
The following antibodies were purchased from Cell Signaling Technology, Inc (Danvers, MA): antimouse immunoglobulin G, horseradish peroxidase-linked antibody, catalog no. 7076, concentration, 1:2000; and antirabbit immunoglobulin G, horseradish peroxidase-linked antibody, catalog no. 7074, concentration, 1:5000.
Primers
The following primers were purchased from QIAGEN (Valencia, CA): RT² quantitative polymerase chain reaction (qPCR) primer assay for human plasminogen activator, urokinase, also known as uPA, catalog no. PPH00796C-200, Entrez Gene identification 5328; RT² qPCR primer assay for human serpin peptidase inhibitor, clade E, also known as PAI-1, catalog no. PPH00215F-200, Entrez Gene identification 5054; RT² qPCR primer assay for human glyceraldehyde-3-phosphate dehydrogenase, catalog no. PPH00150F-200, Entrez Gene identification 2597; and RT² qPCR primer assay for human β-actin (actin, beta), also known as β-actin, catalog no. PPH00073G-200, Entrez Gene identification 60.
Western blot for uPA, PAI-1, and PPAR-γ expression and p38 MAPK phosphorylation
After treatment for 72 hours, the media were removed from cells and a lysis buffer (Cell Signaling Technology) containing 50 mM Tris at pH 7.4, 50 mM NaCl, 1% Triton X-100, 0.1% sodium dodecyl sulfate, 0.3 mM Na-orthovanadate, 50 mM NaF, 1 mM dichlorodiphenyl-trichloroethane, 10 μg/mL leupeptin, and 5 μg/mL aprotinin was added to the cells. Cells were scraped and put into tubes. Protein concentrations were determined by a bicinchoninic assay reagent (Pierce, Rockford, IL).
An equal amount of protein in each sample was separated using NuPAGE Novex 4-12% Bis-Tris gels (Invitrogen) and transferred to nitrocellulose membranes. Membranes were blocked in 5% milk and probed with uPA (Santa Cruz Biotechnology), PAI-1 (Santa Cruz Biotechnology), PPAR-γ (Santa Cruz Biotechnology), phospho-p38 (Santa Cruz Biotechnology), p38α/β (Santa Cruz Biotechnology), and β-actin (Santa Cruz Biotechnology) antibodies. After incubation with the corresponding secondary antibody, proteins were visualized with a chemiluminescence detection system (Pierce). The intensity of the bands was determined using ImageQuant LAS 4000 (GE Healthcare, Life Sciences, Indianapolis, IN).
The expression of uPA, PAI-1, PPAR-γ, and p38 MAPK phosphorylation protein was quantified by a densitometry analysis using Image J software (National Institutes of Health, Bethesda, MD) in which the target protein (uPA, PAI-1, or PPAR-γ) is normalized to a structural protein (β-actin) to control between groups and ensures correction for the amount of total protein on the membrane (phospho-p38 is normalized to total p38).
These replicative glucose exposure experiments were performed 8 times with uPA without inhibitor pretreatment, 8 times with pretreatment using p38 inhibitor, and 4 times with pretreatment using rosiglitazone. For PAI-1 there were 14 replicates without inhibitors, 6 with pretreatment using p38 inhibitor, and 4 with pretreatment with rosiglitazone. The glucose series was repeated 5 times for PPAR-γ and 10 times for p38 MAPK phosphorylation. Sample sizes for these replicates are shown in Figure 1 with representative immunoblots.
Quantitative PCR for uPA and PAI-1
After the treatment, the media were removed from cells, and a lysis/binding solution from the Ambion RNAqueous-4PCR kit (Invitrogen) was added to the cells. Cells were scraped and put into tubes. An equal amount of 64% ethanol was added to the tubes and mixed. The mixed solution was added to a filter and centrifuged into another tube. The filter was then washed with wash solution #1 from the kit and then wash solution #2/#3 from the kit. Preheated elution solution from the kit was added to the filter to elute the ribonucleic acid (RNA) into tubes. Then 10 μL of 10 times deoxyribonuclease (DNAse) I buffer and 1 μL of DNAse I from the kit was added to the RNA tubes.
Samples were incubated in a heat block at 37°C for 30 minutes. Then 11 μL of the DNAse inactivation reagent from the kit was added to the samples and mixed for 2 minutes and then centrifuged. A Nanodrop (Thermo Fisher Scientific, Wilmington, DE) was used to get the concentration of RNA in each sample. Then the sample tubes were heated for 3 minutes at 75°C and put on ice. In a new tube, 2 μL of oligo(deoxythymidine) from the Ambion RETROscript first-strand synthesis kit (Invitrogen) was added to 10 μL of RNA. Then 2 μL of 10 times reverse transcriptase buffer from the kit, 4 μL of deoxynucleotide triphosphate mix from the kit, 1 μL ribonuclease inhibitor from the kit, and 1 μL Molony-murine leukemia virus reverse transcriptase from the kit were added to the tubes for a total volume of 20 μL.
Tubes were mixed gently and spun down. Tubes were placed in a Perkin Elmer GeneAmp 9600 PCR thermal cycler and set to incubate at 42-44°C for 1 hour and then incubate at 92°C for 10 minutes to inactivate the reverse transcriptase. The Nanodrop was then used to get the concentration of complementary deoxyribonucleic acid in each sample. For the real-time polymerase chain reaction (PCR), the primers used were uPA, serpin peptidase inhibitor, clade E (PAI-1), glyceraldehyde-3-phosphate dehydrogenase, and ACTB (β-actin), all purchased from QIAGEN. Also used was iTaq SYBR Green Supermix with ROX (Bio-Rad Laboratories, Hercules, CA). Each well of the PCR plate contained 1 μL of the complementary deoxyribonucleic acid, 1 μL of the primer, 12.5 μL of SYBR Green, and 10.5 μL of water. Quantitative PCR was performed on a Bio-Rad iCycler iQ5 (Bio-Rad Laboratories), using a 2-step cycling program.
Results were analyzer using LinRegPCR software from Dr J. M. Ruijter (The Heart Failure Research Center, Amsterdam, The Netherlands). The replicated glucose exposure experiments were performed 4 times with uPA mRNA without inhibitor pretreatment and 4 times with each inhibitor pretreatment. Similarly, qPCR for PAI-1 mRNA was replicated 4 times without inhibitors and 4 times using each inhibitor pretreatment.
Enzyme-linked immunosorbent assay for a sEng, sFlt-1, VEGF 165, PlGF, and IL-6
After the treatment, the media removed from the cells were place in tubes. Levels of antiangiogenic (sEng, sFlt-1) and angiogenic (VEGF 165, PlGF) factors as well as the level of inflammatory cytokine IL-6 were measured in the culture media by the commercially available enzyme-linked immunosorbent assay (ELISA) kits from R&D Systems (Minneapolis, MN). For sENG, a human endoglin/CD105 quantikine ELISA kit was used. For sFLT-1, a human soluble VEGF R1/Flt-1 quantikine ELISA kit was used. For VEGF 165, a human VEGF quantikine ELISA kit was used. For PlGF, a human PlGF quantikine ELISA kit was used. For IL-6, a human IL-6 quantikine ELISA kit was used. The replicated glucose exposure experiments were performed 4 times for each factor.
Statistical methods
Data are expressed as mean ± SE. Statistical significance is assessed by analysis of variance and a Duncan’s post-hoc test for differences between glucose effects and inhibitor treatments, with P < .05 taken as significant. While interpreting the grafts, asterisks and different letters will signify a statistical significance.
Results
Hyperglycemia up-regulated p38 MAPK phosphorylation and PPAR-γ expression
Figure 2 demonstrates that the ratio of phosphorylated p38 MAPK to the nonphosphorylated p38α/β was significantly ( P < .05) up-regulated in all the CTB cell cultures treated with 495 mg/dL or more of glucose compared with basal (45 mg/dL). Intermediate levels of glucose exposure, beginning at more than 225 mg/dL, produced intermediate changes that were not different from values at either exposure to 45 mg/dL or 495 mg/dL. In addition, Figure 2 shows that PPAR-γ expression was significantly up-regulated in 135 mg/dL or more glucose-treated CTB cells compared with basal (45 mg/dL).
Hyperglycemia down-regulated uPA protein and mRNA expression
Graphs in Figure 3 show that uPA protein and mRNA levels in CTB cells are altered by increasing concentrations of glucose in the culture medium. Glucose levels of 225 mg/dL or greater decreased uPA protein expression compared with basal glucose at 45 mg/dL. However, this decrease was seen at a lower level of 135 mg/dL in uPA mRNA levels. Pretreatment with either a p38 inhibitor or rosiglitazone attenuated these changes.
