醛固酮和皮质醇对胎盘生长的调控-----内分泌杂志

Regulation of Placental Growth by Aldosterone and Cortisol Carine Gennari-Moser, Eliyahu V. Khankin, Simone Schu¨ller, Genevie`ve Escher, Brigitte M. Frey, C.-Bettina Portmann, Marc U. Baumann, Andrea D. Lehmann, Daniel Surbek, S. Ananth Karumanchi, Felix J. Frey, and Markus G. Mohaupt Departments of Nephrology/Hypertension (C.G.-M., S.S., G.E., B.M.F., F.J.F., M.G.M.) and Obstetrics and Gynecology (C.-B.P., M.U.B., D.S.), University Hospital Bern, Inselspital, 3010 Berne, Switzerland; Division of Vascular and Molecular Medicine (E.V.K., S.A.K.), Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts 02215; and Division of Histology (A.D.L.), Institute of Anatomy, University of Bern, 3000 Berne, Switzerland During pregnancy, trophoblasts grow to adapt the feto-maternal unit to fetal requirements. Al- dosterone and cortisol levels increase, the latter being inactivated by a healthy placenta. By con- trast, preeclamptic placental growth is reduced while aldosterone levels are low and placental cortisol tissue levels are high due to improper deactivation. Aldosterone acts as a growth factor in many tissues, whereas cortisol inhibits growth. We hypothesized that in preeclampsia low aldo- sterone and enhanced cortisol availability might mutually affect placental growth and function. Proliferation of cultured human trophoblasts was time- and dose-dependently increased with aldosterone (P�0.04 toP�0.0001) and inhibitedby spironolactoneandglucocorticoids (P�0.01). Mineralo- and glucocorticoid receptor expression and activation upon agonist stimulation was verified by visualization of nuclear translocation of the receptors. Functional aldosterone defi- ciency simulated in pregnant mice by spironolactone treatment (15 �g/g body weight/day) led to a reduced fetal umbilical blood flow (P� 0.05). In rat (P� 0.05; R2� 0.2055) and human (X2� 3.85; P � 0.0249) pregnancy, placental size was positively related to plasma aldosterone. Autocrine production of these steroid hormones was excluded functionally and via the absence of specific enzymatic transcripts for CYP11B2 and CYP11B1. In conclusion, activation of mineralocorticoid receptors by maternal aldosterone appears to be required for trophoblast growth and a normal feto-placental function. Thus, lowaldosterone levels andenhanced cortisol availabilitymaybeone explanation for the reduced placental size in preeclampsia and related disorders. (Endocrinology 152: 0000–0000, 2011) In human pregnancy, appropriate trophoblast implanta-tion and proliferation is a key determinant for fetal and maternal outcome (1). Many factors, among them cyto- kines, other regulatory peptides, and steroid hormones, contribute to trophoblast proliferation (2–4). Compro- mised adaptive growth and proliferation of trophoblasts present major risk factors to develop preeclampsia (5–7). The importance of aldosterone and activation of its respective effector, the mineralocorticoid receptor (MR), appeared in animals departing salt water for a dry and salt-deprived environment. Thus, the first andmost prom- inent role of aldosterone was attributed to the mainte- nance of volume homeostasis by reabsorbing sodium. More recently, evidence was presented that aldosterone directs cell proliferation in several organs including fibro- blasts in the heart, mesangial cells in the kidney, or endo- thelial cells in blood vessels (8, 9). Aldosterone concentrations rise in the luteal phase of themenstrual cycle and, if a conceptionoccurs, increaseby a factor of 20 throughout pregnancy toward term (10). It is believed that this is an adaptive mechanism to expand the plasma volume allowing for an appropriate utero-pla- ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2011 by The Endocrine Society doi: 10.1210/en.2010-0525 Received May 11, 2010. Accepted October 1, 2010. Abbreviations: FBS, Fetal bovine serum;GR,glucocorticoid receptor;MR,mineralocorticoid receptor; TLC, thin-layer chromatography. R E P R O D U C T I O N - D E V E L O P M E N T Endocrinology, January 2011, 152(1):0000–0000 endo.endojournals.org 1 Endocrinology. First published ahead of print November 10, 2010 as doi:10.1210/en.2010-0525 Copyright (C) 2010 by The Endocrine Society cental perfusion (11). In preeclampsia, for reasons only partially understood, plasma volume and placental size are reduced, whereas aldosterone levels are low, suggest- ing a new functional relationship (12–16). Total cortisol concentrations increase in normal preg- nancy.Theaccessof cortisol to theglucocorticoid receptor (GR) within the placenta is intracellularly controlled by the enzyme 11�-hydroxysteroid dehydrogenase type 2 (17). This enzyme converts biological active cortisol into inactive cortisone. In preeclampsia, reduced activity of the 11�-hydroxysteroid dehydrogenase type 2 exposes tro- phoblasts to increased cortisol concentrations, whereas we demonstrated earlier that cortisol is virtually absent in normal placentas (18–21). Several lines of evidence, in- cluding recent data from our group, support the assump- tion that placental and fetal development is impaired by a high exposure to glucocorticoids (21, 22). Interestingly, the placental size was below average in adrenalectomized ewes supplemented with cortisol in the absence of aldosterone, whereas substitution with either aldosterone or aldosterone combined with cortisol re- sulted in a normal placental size (23). Although the former studies strongly suggest a role for aldosterone in placental growth, the underlying mechanism remained to be estab- lished. Thus, it is conceivable that the aldosterone-de- pendent changes in plasma volume determine placental growth. Alternatively, aldosterone might directly stimu- late placental cell proliferation. Because cortisol cannot replace aldosterone for the normal development of the placenta and a high cortisol availability appears to be det- rimental for placental growth, the assumed antagonistic interplay between aldosterone and cortisol for the pla- centa is intriguing. We hypothesize now that aldosterone directly stimu- lates placental growth, and that this regulation is opposed by cortisol. Thus, high aldosterone concentrations would support the feto-maternal unit to cover the requirements during pregnancy by adapting it to environmental condi- tions via a nonrenal effect. Materials and Methods Material and cell lines Collagen I-coated cell culture plates used for primary tropho- blasts were from Becton Dickinson (Basel, Switzerland), those for immortalized cell lines from Techno Plastic Products AG (Trasadingen, Switzerland). Cell culture media were from Life Technologies, Inc./Invitrogen (Basel, Switzerland) except for McCoy’s (Sigma, Buchs SG, Switzerland). Aldosterone, spirono- lactone, mifepristone, dimethyl sulfoxide, cortisol, glycyrrhet- inic acid, dexamethasone, and fetal bovine serum (FBS) were from Sigma. L-glutamine, penicillin/streptomycin, and HEPES were provided by Life Technologies, Inc./Invitrogen. Penicillin, streptomycin, and amphotericin B used for primary cell cultures was obtained from Invitrogen (Basel, Switzerland). Insulin- transferrin-seleniummix, and supplemented 25%FBS in Ham’s F12 medium base were obtained from Becton Dickinson. 3H- thymidine (specific activity 70–90 Ci/mmol) was purchased from PerkinElmer (Boston, MA), whereas 3H-corticosterone (specific activity 70 Ci/mmol) was from Amersham (Bucking- hamshire, UK) and 3H-deoxycorticosterone (specific activity 40–60 Ci/mmol) from American Radiolabeled Chemicals (St. Louis,MO).Thin-layer chromatography (TLC)plates (silica gel) were from Macherey-Nagel (Du¨ren, Germany). For immunofluorescence studies, the following antibodies were used: antivimentin (monoclonal mouse, clone V9; Dako, Baar, Switzerland), anticytokeratin-7 (monoclonal mouse; Sigma), anti-MR and anti-GR (rabbit polyclonal; Santa Cruz Biotechnology, Inc.,Heidelberg,Germany). Their respective flu- orescence-labeled secondary antibodies (antimouse and antirab- bit,AlexaFluor)wereobtained fromInvitrogen, andVectashield mountingmediumwas acquired fromVector Laboratories (Bur- lingame, CA). Rat tail collagen I was from Roche Diagnostics GmbH (Mannheim, Germany). Thehumanchoriocarcinoma JEG-3, the kidneyAfricanmon- key COS-7, and the human adrenal H295R cells were obtained from American Type Culture Collection (Manassas, VA). The human first-trimester trophoblast cell line HTR-8/SV neo was a gift from Charles H. Graham (Queen’s University, Kingston, Ontario, Canada). Human extravillous primary trophoblasts were isolated from first-trimester (gestational wk 7–12) tissue of aborted fetuses of healthy donors after informed consent and approval of the local institutional review board. In detail, villi from the fetal part of the placenta were carefully removed by forceps, collected in 1� PBS (Inselspital Hospital Pharmacy, Berne, Switzerland) and spun down at 1500� g for 5 min. After aspirating the supernatant, the pellet was resuspended in 20 ml 0.05% trypsin-ethylenediamine tetra-acetate and incubated for 10minat37C.The resulting suspensionwasmixedwithaplastic pipette and filtered through a 70-�m cell strainer (Falcon/BD Biosciences, Basel, Switzerland). The cell strainer was washed twice with cell culture medium. The filtrate was centrifuged at 1500 � g for 5 min to pelletize the cytotrophoblasts. The cy- totrophoblasts were resuspended in DMEM/F12 (Life Technol- ogies, Inc., Basel, Switzerland) containing 10% FBS, glutamax, penicillin, streptomycin, and amphotericin B and plated on col- lagen I-coated culture dishes. Cells were cultured and passaged given subconfluence at 37 C and 5% CO2. Characterization and functional assessment of trophoblasts by immunofluorescence and confocal laser microscopy Human first-trimester primary trophoblasts were charac- terized after being cultured in standard conditions on collagen I-coated coverslips for 24 h in DMEM/F12 supplemented with 10% FBS. At room temperature, cells were washed twice with PBSand fixed in4%formaldehyde for15min.Thiswas followed by repeated washing with PBS before and after the addition of NH4Cl (50mM) for 15min. Cellmembraneswere permeabilized withTritonX-100 (0.1%) for 5min andwashedwith PBSbefore blocking in 10% FBS for 30 min. Incubation with primary an- tibodies against vimentin and cytokeratin-7 (dilution 1:100 in 1% FBS/PBS) was performed for 1 h at room temperature. After washing with PBS, cells were incubated with the secondary an- 2 Gennari-Moser et al. Aldosterone, Cortisol, and Placental Growth Endocrinology, January 2011, 152(1):0000–0000 tibody (antimouse, 1:200 in 1% FBS) in the dark for 30 min. After repeated washes with PBS, coverslips were analyzed with a confocal laser scanningmicroscope [Zeiss LSM510Metawith inverted Zeiss microscope Axiovert 200M and Ar (488 nm) and diode laser (405 nm) with a Plan-Apochromat 63�/1.4 objective; ZeissFeldbach,Switzerland] followingstandardprocedures. Image processing and visualization was done using IMARIS, a three-di- mensionalmultichannel imageprocessingsoftwareforconfocalmi- croscopic images (Bitplane AG, Zurich, Switzerland). JEG-3, HTR-8/SV neo, H295R, COS-7 cells, as well as hu- man first-trimester primary trophoblasts, were cultured on cov- erslips for 24 h in their corresponding medium supplemented with up to 10% FBS as indicated. To characterize functional activity of GR and MR, cells were washed with PBS and re- mained either unstimulated or were exposed for 40min to 2 h to cortisol (10�7 M), aldosterone (10�7 M), and in combination with the respective antagonists RU486 (10�6 M) or spironolac- tone (10�6M), respectively, in serum-freemedium.To ensure the specificity of the antibodies, the aldosterone (plasmid gifted by Dr. S. Rusconi, University of Fribourg, Switzerland) or the GR (plasmid gifted by Dr. A. Odermatt, University of Basel, Swit- zerland) were expressed in COS-7 cells. We only identified a positive staining using antibodies against the respective receptor (see figure 3A). Negative controls remained negative. CYP11B1 and CYB11B2 activity assay in intact cells Human first-trimester primary extravillous cytotrophoblasts, JEG-3,HTR-8/SV neo, COS-7, andH295Rwere cultured for 24 h in their correspondingmedium supplementedwith either 5 or 10% FBS or 5%of supplemented 25%FBS inHam’s F12mediumbase. After 24 h, cells were washed with PBS, and fresh serum-free me- dium or medium containing 1% FBS supplemented with 1 �Ci/ml 3H-corticosterone or 3H-deoxycorticosterone was added for 8 to 24 h. Medium was collected and processed as described above, except that steroids were resuspended in 30 �l nonradioactive corticosterone/aldosterone or deoxycorticosterone/corticoste- rone/aldosterone mixture (10 mg/ml). TLC was performed in a chamber saturated with dichloromethane:methanol:H2O (150: 10:1; vol:vol:vol). 3H-corticosterone, 3H-deoxycorticosterone, and 3H-aldosterone were detected, counted, and used to calcu- late enzyme activity. Using this method, apparent enzyme activ- ities of CYP11B2were calculated as aldosterone/(corticosterone� aldosterone) * 100andasaldosterone/(deoxycorticosterone�cor- ticosterone � aldosterone) * 100. Assays were performed in trip- licates. In a separate set of experiments,we added 3H-progesterone to the cells and determined the conversion to either cortisol, deoxy- corticosterone, 11-deoxycortisol, corticosterone, and aldosterone. Real-time PCR Extraction of total RNA was performed using the SV total RNA isolation kit (Promega, Du¨bendorf, Switzerland). RNA was reverse transcribed (Promega). For real-time PCR, primers and probes for CYP11B1, and CYP11B2, were obtained from Applied Biosystems (CYP11B1, Hs01596404_m1; CYP11B2, Hs01597732_m1) (Applied Biosystems, Foster City, CA). The 18S signal served as endogenous control. Assessment of proliferation by 3H-thymidine proliferation assay and cell counting JEG-3, HTR-8/SV neo, and human first-trimester primary extravillous cytotrophoblasts were cultured for 24 h, washed with PBS, and starved for 24 h to synchronize the cells. After repeated washes with PBS, the cells were incubated for up to 120 h using media conditions specific to the experimental pro- cedure. Experiments includedadditionof either aldosterone, spi- ronolactone, cortisol, RU486, glycyrrhetinic acid, dexametha- sone, and/or combinations hereof with the respective solvents at the indicated concentrations. At the end of the experiments, 1 �Ci 3H-thymidine was added per ml medium, and the cells were incubated for 6 h at 37 C. Medium was removed, cells were washed twice with ice-cold PBS, and 500 �l of 10% ice-cold trichloroacetic acid were added per well. DNA was precipitated for 30 min at 4 C, TCA removed, washed with PBS, and 250 �l lysis buffer containing 0.5 N NaOH and 0.5% sodium dodecyl sulfate was added per well. After a 15-min incubation at room temperature, 100�l were transferred into scintillation vials con- taining 4 ml Irgasafe, and 3H-thymidine incorporation was measured. In addition, cell number was counted after removing the me- dium, washing the cells with PBS, harvesting and staining with trypan blue by using a Neubauer chamber. In vivo assessment of the role of aldosterone in animal models and human pregnancy All animal studies were approved by the appropriate institu- tional animal care and use committee. Pregnant mice arrived at d 5 of gestation and were allowed to recover for 1 d. Mice had free access to water and standard chow (containing 0.42% NaCl). On gestational d 6, Alzet osmotic minipumps providing a continuous spironolactone delivery of 0.625 �g/g body weight/hwere sc implanted under general anesthesiawith isoflu- rane (Durect Corp., Cupertino CA) (24). On gestational d 18, umbilical vein perfusion was measured using ultrasound Dopp- ler technique to assess blood velocity (VeVo 770; Visual Sonics, Toronto, Canada) in general anesthesia with isoflurane. Sprague Dawley rats (n� 18) arrived at d 5 of gestation and were allowed to recover for 1 dwith free access to standard chow (containing 0.42% of NaCl) and water. Aldosterone levels were measured in serum samples obtained at d 6 and 18 in animals under general anesthesia with isoflurane by ELISA (Cayman Chemicals, Ann Arbor, MI) according to manufacturer specifica- tions. Animalswere killed at d 18by exsanguination, andplacental wet weight was obtained after removal of hydramnial fluid. A protocol to study pregnant women, fulfilling the criteria of the Declaration of Helsinki, was approved by the local ethical review board. All participating women gave informed consent. These pregnant womenwere recruited for a first-trimester blood sampling, including aldosterone plasma concentration taken in a supine position. Placental weight was taken shortly after de- livery. Nonsingleton births were excluded. Placental weight was corrected for gestational week, sex of the child, and mode of delivery (25–27). Mean age of the women (n � 34) was 32.4 � 1.0 yr, bodymass index 26.8� 1.7 kg/m2, neonatal birthweight 2900 � 159 g, gestational age 37.7 � 0.5 wk, and all mothers were normotensive with a blood pressure of systolic 117.0� 2.8 mm Hg and diastolic 76.1 � 2.3 mm Hg. Statistical analysis All data arepresentedasmean� SD.Todetermine statistically significant differences between more than two groups, one-way ANOVA was used, followed by Bonferroni or Dunnett’s post hoc tests for multiple comparisons. Unpaired t test was used to Endocrinology, January 2011, 152(1):0000–0000 endo.endojournals.org 3 analyze the difference observed between two groups. For com- parison of placental percentile distributions, theX2 and Fisher’s exact test were used, taking the absence of a placental percentile less than 25% corrected for gestational week as threshold. Sig- nificance was assigned at P� 0.05. All statistical analyses were performed using SYSTAT version 12 (SPSS, Inc., Chicago, IL). Significance levels are categorized in all figures as **, P � 0.01 and *, P � 0.05 nonsignificant. Results Characterization of the trophoblast cell lines JEG-3 and HTR-8/SV neo and of human first-trimester primary trophoblasts in culture We determined the expression of vimentin and cyto- keratin-7 to characterize the trophoblasts by immuno- fluorescence. Proliferative response of the trophoblast cell lines JEG-3 and HTR-8/SV neo and human first-trimester primary trophoblasts to aldosterone in culture After establishing baseline conditions to maintain the cultured cells at the lowest FBS concentrations necessary, the effect of aldosterone on trophoblast growth was in- vestigated. In JEG-3 cells, aldosterone (10�7 M) exerted a proliferative effect at 48 h if added in the absence of FBS (P � 0.0001 vs. control). In similar conditions, cultured HTR-8/SV neo trophoblast cells also showed a slight, but consistent, growth response to aldoste- rone (10�7M) at 72 h (P�0.008 vs.0% FBS control). In human first-trimester primary trophoblasts, the proliferation in the presence of aldosterone (10�7 M) and 1% FBS already peaked at a 24-h incubation in independent cell isolations spanning several gestational weeks within the first trimester (P � 0.0002 and P � 0.0097 vs. 1% FBS control conditions for isolations at gestational wk 74/7 and 10 1/7, respectively), which returned to baseline after 48 h. To ex- clude polyploidy of trophoblasts, pro- liferation determined by 3H-thymidine incorporation was verified by trypan blue staining and cell counting. In addi- tion, we microscopically analyzed and excluded polyploidism. Aldosterone dose-dependently stim- ulated growthof primary first-trimester human trophoblasts maintained in se- rum-free conditions and even in cell cul- ture medium containing 10% FBS (Fig. 1A). Proliferationdue to aldosterone ei- ther assessedbycell countingor 3H-thy- midine incorporation (Fig. 1B) was dose-dependentl