Microvesicles and exosomes released by amnion epithelial cells under oxidative stress cause inflammatory changes in uterine cells^†

Abstract

Extracellular vesicles play a crucial role in feto-maternal communication and provide an important paracrine signaling mechanism in pregnancy. We hypothesized that fetal cells-derived exosomes and microvesicles (MVs) under oxidative stress (OS) carry unique cargo and traffic through feto-maternal interface, which cause inflammation in uterine cells associated with parturition. Exosomes and MVs, from primary amnion epithelial cell (AEC) culture media under normal or OS-induced conditions, were isolated by optimized differential centrifugation method followed by characterization for size (nanoparticle tracking analyzer), shape (transmission electron microscopy), and protein markers (western blot and immunofluorescence). Cargo and canonical pathways were identified by mass spectroscopy and ingenuity pathway analysis. Myometrial, decidual, and cervical cells were treated with 1 × 10⁷ control/OS-derived exosomes/MVs. Pro-inflammatory cytokines were measured using a Luminex assay. Statistical significance was determined by paired T-test (P < 0.05). AEC produced cup-shaped exosomes of 90–150 nm and circular MVs of 160–400 nm. CD9, heat shock protein 70, and Nanog were detected in exosomes, whereas OCT-4, human leukocyte antigen G, and calnexin were found in MVs. MVs, but not exosomes, were stained for phosphatidylserine. The protein profiles for control versus OS-derived exosomes and MVs were significantly different. Several inflammatory pathways related to OS were upregulated that were distinct between exosomes and MVs. Both OS-derived exosomes and MVs significantly increased pro-inflammatory cytokines (granulocyte-macrophage colony-stimulating factor, interleukin 6 (IL-6), and IL-8) in maternal cells compared with control (P < 0.05). Our findings suggest that fetal-derived exosomes and MVs under OS exhibited distinct characteristics and a synergistic inflammatory role in uterine cells associated with the initiation of parturition.

Oxidative stress-induced fetal membrane cells produce exosomes and MVs with distinct properties and cargo and may function as paracrine signalers at the feto-maternal interface during pregnancy and parturition.

Introduction

Extracellular vesicles (EVs) are nano-sized membrane-bound vesicles released by different cell types [1]. The nomenclature of EVs refers to a heterogeneous population that differs in size, biogenesis, composition, and function [2]. They are usually divided into different classes, of which the most important are exosomes (40–160 nm), microvesicles (MVs, 150–1000 nm), and apoptotic bodies (>1000 nm) [3]. However, the physical differences between exosome and MV classes are subtle [4]. The role of exosomes in cell-to-cell communication and their functional roles in facilitating both physiologic and pathologic processes have been widely studied; however, little is known about the content, surface markers, and function of MVs.

MVs, also known as “ectosomes,” “microparticles,” or “shedding vesicles,” are large vesicles shedding from the surface of various cells [5]. In addition to some differences in their sizes and surface markers, the major difference between exosomes and MVs is in their biogenesis and release [2]. MVs are released by outward budding and shedding from the plasma membrane, whereas exosomes are formed by inward budding of part of endosomes during their conversion to multivesicular bodies and the generation of intraluminal vesicles [5]. Exosomes are then released from the cells by exocytosis. The release of exosomes and MVs is governed by several endosomal sorting complex required for transport (ESCRT) machinery and cytoplasmic proteins [2, 6]. MVs have been identified in several body fluids that demonstrate their role in normal and diseased conditions [7–11]. This suggests the potential role of MVs as a communication channel, therapeutic target, and diagnostic tool for many diseases, thus encouraging continued research.

During pregnancy, the number of EVs increases significantly and they become active mediators for feto-maternal crosstalk [12]. Besides maternal tissues, EVs are also released from the placenta and fetal tissues for communication [13, 14]. The role of EVs during normal and complicated pregnancies has been widely studied. It has been demonstrated that EVs play a vital role during implantation, regulating endometrial remodeling, and immunomodulation through altering inflammatory responses at different stages, and initiating labor and parturition [15–20]. In addition, the changes in the concentration of EVs, content, and bioactivity have been accompanied by several pregnancy-associated disorders such as preeclampsia, gestational diabetes, and preterm delivery [12].

Our laboratory is focused on fetal-derived EVs and their role in parturition at term and preterm. Several studies from our laboratory have demonstrated signal trafficking through exosomes from fetal membranes (the innermost lining of the uterine cavity) that can perform paracrine functions [21–23]. Exosomes released from senescent (aging) fetal tissues at term carry inflammatory mediators that could transition the quiescent maternal tissues to an active phase of labor [14, 24, 25]. In vitro and in vivo studies demonstrated the functional role of senescent amnion epithelial cells (AEC)-derived exosomes in triggering parturition [13, 23, 24, 26]. However, there are no reports about the function of AEC-derived MVs and how they differ from exosomes in their contribution to parturition. Therefore, we hypothesized that MVs released from fetal membrane cells carry unique cargo that differs from exosomes and produces a distinct functional effect on maternal cells.

To test our hypothesis, exosomes and MVs from human AECs grown under normal and oxidative stress (OS) environments (to mimic term parturition) were isolated and characterized, cargo differences were determined, and their functional properties in eliciting an inflammatory response in maternal myometrial, decidual, and cervical epithelial cells were tested. Understanding these differences and similarities between these two EVs populations will help in better understanding the mechanisms of feto-maternal paracrine signaling that can be associated with labor triggers. In addition, data from these studies may indicate the usefulness of exosomes and MVs as potential biomarkers to predict pregnancy complications.

Materials and methods

Institutional review board (IRB) approval—no subjects were recruited or consented for this study. Placental specimens used for this study were deidentified and considered as discarded human specimens that do not require IRB approval. Placental specimens were collected from John Sealy Hospital at the University of Texas Medical Branch at Galveston, Texas, USA, in accordance with the relevant guidelines and regulations of approved protocols for various studies.

Isolation and culturing of primary AECs from human placenta

Placenta from pregnant women at term (37–41 weeks’ gestation) prior to the onset of labor were collected. Exclusion criteria included a history of preterm labor and delivery, premature rupture of the membranes, preeclampsia, placental abruption, intrauterine growth restriction, gestational diabetes, group B streptococcus carrier status, history of treatment for urinary tract infection, sexually transmitted diseases during pregnancy, chronic infections such as HIV and hepatitis, and history of cigarette smoking or reported drug and alcohol abuse.

The fetal membrane was separated from the placenta and AECs were isolated from the amniotic membrane as previously described [25]. Briefly, the amniotic membrane was cut into small pieces of ~2 × 2 cm. A digestion buffer composed of 0.25% trypsin and 0.125% collagenase A (Sigma Aldrich) in Hank balanced salt solution (Mediatech Inc., Manassas, VA) was used for double digestion of the amniotic membrane by incubation at 37°C for 35 min with shaking every 15 min. The digestion buffer was inactivated with an equal volume of complete Dulbecco-Modified Eagle Medium (DMEM):nutrient mixture F-12 (DMEM/F12; Mediatech Inc.) supplemented with 1% amphotericin B, 1% penicillin/streptomycin (Mediatech Inc.), 10 ng/mL epidermal growth factor (Sigma Aldrich), and 10% exosome-free heat-inactivated fetal bovine serum (FBS; Sigma Aldrich). Exosome-free FBS was prepared by ultracentrifugation at 100 000 g for 16 h, and then the supernatant was collected and sterilized by filtration through a 0.22 μm Steriflip filter unit (Millipore, Billerica, MA).

The cell suspension was then filtered through a 70 μm cell strainer (Fisher Scientific, Waltham, MA). The collected cell filtrate was centrifuged at 3000 rpm for 10 min, the supernatant was discarded, and the cell pellet was resuspended in a suitable volume of complete DMEM/F12 media. Cells were cultured in a T75 flask (~3 million cells/flask) and incubated at 37°C in humidified 5% carbon dioxide (CO₂) until they reached 70–80% confluence. The purity of AEC verified by immunocytochemistry using anti-human cytokeratin antibodies as previously prescribed [27–30], and all our cultures had >95% cytokeratin-positive cells.

Culturing AECs under OS-induced condition

OS was induced in cells by using cigarette smoke extract (CSE) as previously discussed [25]. A stock of CSE was prepared by infusing the smoke of a single commercial cigarette into 25 mL of exosome-free complete DMEM/F12 medium. The stock CSE was diluted 1:50 in exosome-free complete DMEM/F12 media before use. AECs were cultured until reaching 70–80% confluence, and then the cultured flask was rinsed with sterile 1X phosphate-buffered saline (PBS) followed by treatment with the cell media (control conditions) or diluted CSE-containing cell media (OS conditions). Cells were incubated at 37°C in 5% CO₂ and 95% air humidity. After 48 h, culture media were collected, and the total cell numbers/flask were counted using a hemocytometer. The culture media from both control and OS treatments were stored at −80°C.

Exosomes and MVs isolation from AEC culture media

The frozen culture media was thawed overnight at 4°C. The isolation of EVs was carried out by using differential centrifugation and ultracentrifugation method, as previously described with some modifications (Figure 1) [22, 26]. The culture media were sequentially centrifuged at 300 g for 10 min and at 2000 g for 20 min to remove any cell debris using a Sorvall Legend X1R and TX-400 swinging bucket rotor (Thermo Fisher Scientific). The supernatant was collected and transferred to an Amicon ultra-15 100 KDa device to concentrate it down to 2 mL by centrifugation at 4000 g for 30 min. The concentrated sample was then collected from the collection device (~200–300 μL) and transferred to a microcentrifuge tube. The Amicon tube was washed with 200 μL cold PBS to rinse each filter tube, and then transferred to the same microcentrifuge tube. The collected sample was filtered through a 0.8 μm filter. Afterward, the microcentrifuge tubes were centrifuged at 10 000 g for 30 min. The pellets and supernatant were collected, and the pellets were washed by resuspending in 1 mL PBS, recentrifuging at 10 000 g for 30 min to collect MVs. The supernatant was filtered through 0.22 μm Nalgene Syringe Prefilter Plus filters and ultracentrifuged (Beckman Optima LX-80 ultracentrifuge, 70.1Ti rotor, Beckman Coulter) at 100 000 g for 2 h to collect exosomes. The supernatant was discarded, whereas the pellet was resuspended in 100 μL PBS and then passed through an Exo-spin column (Cell Guidance System LLC, MO, USA). Purified exosomes were eluted by the addition of 200 μL PBS to the column. Collected exosomes and MVs suspensions were stored at −80°C until analysis.

Figure 1.

Schematic diagram showing the isolation steps of microvesicles and exosomes by the optimized differential centrifugation method.

Characterization of EVs by nanoparticle tracking analysis (NTA)

The size and concentration of isolated control exosomes and MVs were analyzed using a Nanosight instrument (NS300) equipped with an sCMOS camera and a 405 nm laser at a temperature of 25°C. Exosomes and MVs samples suspended in PBS were diluted with filtered Milli-Q water (Millipore Sigma, St. Louis, MO) and injected into the Nanosight instrument. A single measurement consists of 30 s videos and three repeats at camera level 12. The detection threshold was set at four and data acquisition and processing were performed using the NTA software. The average number of particles/mL and size were determined for each sample.

Transmission Electron Microscopy (TEM)

The shape and size of exosomes and MVs were determined using a TEM (Jeol, Peabody, MA). Briefly, 20 μL of exosomes or MVs suspension in PBS was applied onto formvar/carbon-coated 200-mesh copper grids and left to dry at room temperature for 10 min. The EVs-covered grids were negatively stained using 2% aqueous uranyl acetate for 1 min, filtered through a 0.2 μm filter, and left to dry at room temperature. The grid was directly placed into a grid box until observation. The grids were examined using a Philips CM-100 TEM at 60 kV. Images were acquired on a bottom-mounted CCD camera Orius SC2001 (Gatan, Pleasanton, CA). A minimum of five frames were viewed per sample.

Exosome antibody array

A commercially available Exo-Check Exosome Antibody Array kit (System Biosciences, USA) was used to identify the expression of specific protein markers on exosomes and MVs samples according to the manufacturer’s protocol. Each array membrane has eight antibodies (CD63, CD81, ALIX, FLOT1, ICAM1, EpCAM, ANXA5, and TSG101), GM130 as a control for cellular contamination, positive controls containing human serum exosome proteins, and negative control. The membrane was developed with Clarity Western Enhanced chemiluminescence (Bio-Rad, USA) and analyzed using a ChemiDoc Imaging System (Bio-Rad).

Western blot

10X Radioimmunoprecipitation assay buffer (RIPA) lysis buffer (0.50 M Tris pH 8.0, 1.50 M NaCl, 10% Triton X, 5% sodium deoxycholate, and 10% SDS) supplemented with a protease and phosphatase inhibitor cocktail was added to the exosomes and MVs samples suspended in PBS for a final concentration of 1X. The lysis mixture was vortexed for 30 s, sonicated for 30 s, and kept on ice for 30 min. MVs samples were centrifuged at 10 000 g for 20 min at 4°C and the supernatants were collected, whereas exosomes samples were kept without centrifugation. This was done to get rid of any residual contaminant in the MVs samples such as cell debris. The centrifugation step was not needed for the exosomes as we used Exo-spin columns during the isolation step, as previously mentioned, which provides a clean exosomes sample. Protein concentrations in the prepared samples were determined using a Pierce BCA protein assay kit (Pierce, Rockford, IL).

The protein samples (~5–8 μg) were separated using sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) on a gradient (4–15%) Mini-PROTEAN1TGX Precast Gels (Bio-Rad, Hercules, CA) and transferred to the membrane using the Trans-Blot Turbo Transfer System (Bio-Rad, USA). Membranes were blocked in 5% nonfat milk in 1X Tris-buffered saline-Tween 20 (TBS-T) buffer for 2 h at room temperature. Membranes were probed (or reprobed) with the primary antibody overnight at 4°C. The membrane was incubated with a suitable secondary antibody conjugated with horseradish peroxidase. The membranes were developed with Clarity Western Enhanced Chemiluminescence Substrate (Bio-Rad, USA) and analyzed using the ChemiDoc Imaging System (Bio-Rad). For the reprobing of the same membrane, a stripping protocol was implemented by following the instructions for Restore Western Blot Stripping Buffer (Thermo Fisher). Blots were not used for more than three times.

The following anti-human antibodies were used for western blot: exosome markers CD81 (Catalog# MAB6425; Abnova) and CD63 (Catalog# NBP-2-32830; Novus Bio), embryonic stem cell markers Nanog (Catalog# 3580S; Cell Signaling), and octamer-binding transcription factor 4 (OCT; Catalog# ----) (Catalog# 2750S; Cell Signaling) were diluted at 1:400. Human leukocyte antigen G (HLA-G; Catalog# SC-21799; Santa Cruiz) was diluted at 1:500. Heat shock protein (HSP) 70 (Catalog# ab5444; Abcam) was diluted at 1:1000. All primary antibodies were diluted in 5% bovine serum albumin (BSA) in TBS-T.

Staining exosomes and MVs with 3,3'-Dioctadecyloxacarbocyanine lyophilic tracer (DiO) and Fluorescein isothiocyanate (FITC)-conjugated Annexin-V

To differentiate between the EVs subpopulation, exosomes and MVs pellets were resuspended in 100 μL 1X Annexin-binding buffer containing 5 μL FITC-conjugated Annexin-V Alexa Fluor 488 (Invitrogen) and kept rocking for 1 h in the dark at room temperature. Another sample of exosomes was stained by resuspending in 100 μL of 100 μM lipophilic tracer-DiO (D275, Invitrogen) and then incubated at 37°C for 30 min. Afterwards, 300 μL of 5% BSA in PBS was added and the volume was completed to 5 mL with PBS. Excess dye (Annexin-V and DiO) was removed by Amicon-ultra 100 KDa through centrifugation at 4000 g for 10 min, and 50 μL of the stained exosomes/MVs with FITC-conjugated Annexin-V or exosomes with DiO were added to a glass coverslip and left to dry. EVs were imaged using BZ-X810 Keyence fluorescence microscopy.

Proteomic analysis of exosomes and MVs

Protein clean-up and digestion

The samples were prepared as described previously with some modification [14, 31]. Briefly, 4 μL of 0.5 M Tris (2-carboxyethyl) phosphine (Thermo Scientific) was added to 200 μL of exosome and MVs samples (control and CSE) in PBS, followed by incubation at 55°C for 1 h with gentle shaking. The sample was then cooled to room temperature and 8.0 μL of 0.5 M iodoacetamide acid was added and allowed to react for 45 min in the dark. The samples were mixed with five times their volume of cold acetone and incubated overnight at −20°C to denature and precipitate the proteins. On the second day, the samples were centrifuged at 15 000 g for 20 min at 4°C. The supernatant was carefully discarded, another 1 mL of ice-cold acetone was added, and the samples were centrifuged at 2800 g for 15 min at 4°C. After discarding the supernatant, the pellets were dried using a speed vacuum for 5 min. The pellets were reconstituted in 25 μL of 5% SDS, 8 M urea, and 50 mM triethylammonium bicarbonate (TEAB; Thermo Scientific), pH 7.5 by vortexing, followed by incubation at 37°C for 30 min. Then, 2.7 μL of 12% phosphoric acid was added to the 25 μL protein solution, followed by 165 μL of binding buffer (90% methanol, 100 mM TEAB final; pH 7.5). The resulting solution was added to an S-Trap spin column (Protifi, Farmingdale, NY) and passed through the column using a benchtop centrifuge (4000 g for 2 min). The spin column was then washed three times with 150 μL of binding buffer and centrifuged (4000 g for 2 min). Trypsin (Catalog #V5280; Promega, Madison, WI) was then added to the protein mixture in a ratio of 1:25 in 50 mM TEAB and incubated at 47°C for 2 h. Peptides were eluted by centrifugation at 4000 g for 2 min with 45 μL of 50 mM TEAB, followed by 45 μL of 0.2% formic acid, then 35 μL of 50% acetonitrile (ACN)/0.2% formic acid, and finally 35 μL of 80% ACN, 0.1% formic acid. The combined peptide solution was then dried in a speed vacuum (room temperature, 1.5 h) and resuspended in 2% ACN, 0.1% formic acid, and 97.9% water, and aliquoted into an autosampler vial.

Nanoflow liquid chromatography–tandem mass spectrometry (nanoLC–MS/MS) analysis

Peptide mixtures were analyzed by nanoLC–MS/MS using a nano-LC chromatography system (UltiMate 3000 RSLCnano, Dionex, Thermo Fisher Scientific, San Jose, CA). The nanoLC–MS/MS system was coupled online to a Thermo Orbitrap Fusion mass spectrometer (Thermo Fisher Scientific, San Jose, CA) through a nano-spray ion source (Thermo Scientific). A trap-and-elute method was used to desalt and concentrate the sample while preserving the analytical column. The trap column (Thermo Scientific) was a PepMap 100 C18 (100 μm × 20 mm, 5 μm particle size), whereas the analytical column was an Acclaim PepMap 100 (75 μm × 25 cm) (Thermo Scientific). After equilibrating, the column in 98% solvent A (0.1% formic acid in water) and 2% solvent B (0.1% formic acid in ACN), the samples (5 μL in solvent A) were injected onto the trap column and subsequently eluted (300 nL/min) by gradient elution onto the C18 column as follows: isocratic at 2% B, 0–5 min; 2–6% B, 5–6 min; 6–32% B, 6–65 min; 32–50% B, 49–50 min; 50–90% B, 71–72 min; isocratic 90% B, 72–73 min; 90–5%, 73–74 min; isocratic 5% B, 74–74.5 min; 5–90%, 74.5–75 min; isocratic 90% B, 75–76 min; 90–2%, 76–77 min; and isocratic 2% B, 77–90 min.

All LC–MS/MS data were acquired using XCalibur, version 4.7.73.11 (Thermo Fisher Scientific) in positive ion mode utilizing a top-speed data-dependent acquisition method with a 3 s cycle time. The survey scans (m/z 375–1500) were acquired in the Orbitrap at a 120 000 resolution (at m/z = 400) in profile mode, with a maximum injection time of 50 ms and an automatic gain control (AGC) target of 400 000 ions. The S-lens Radio Frequency (RF) level was set to 60. Isolation was performed in the quadrupole with a 1.6 Da isolation window, and Collision-induced dissociation (CID) MS/MS acquisition was performed in profile mode using a rapid scan rate with detection in the ion trap using the following settings: parent threshold = 5000; collision energy = 35%; maximum injection time = 35 ms; AGC target 2000 ions. Monoisotopic precursor selection and charge state filtering were on, with charge states 2–7 included. Dynamic exclusion was used to remove selected precursor ions, with a +/− 10 ppm mass tolerance, for 60 s after the acquisition of one MS/MS spectrum.

Database searching

Tandem mass spectra were extracted, and the charge state was deconvoluted using Proteome Discoverer (Thermo Fisher, version 2.4.1.15). Deisotoping was not performed. All MS/MS spectra were searched against the UniProt Human database (reviewed June 11, 2019) using Sequest. Searches were performed with a parent ion tolerance of 10 ppm and a fragment ion tolerance of 0.60 Da. Trypsin was specified as the enzyme, allowing for two missed cleavages. Fixed modification of carbamidomethyl (C) and variable modifications of oxidation (M) and deamidation (N, Q) were specified in Sequest.

Ingenuity pathway analysis (IPA) of identified proteins

Pathway enrichment analyses were performed with IPA (Qiagen, Hilden, Germany) using Fisher exact test. IPA was performed to identify and compare canonical pathways, diseases, and function for the protein cargo of control and CSE EVs based on fold change and Z-scores. Significantly enriched pathways for the proteins and scenarios were identified using P < 0.01.

Culturing of myometrial, decidual, and cervical cells

The immortalized myometrial cells used for this study were a gift from Prof. Sam Mesiano, Case Western Reserve University, Cleveland, OH. These cell lines were prepared as previously described [25, 32]. Myometrial cells were cultured in a T75 flask containing DMEM 1X (Mediatech Inc., Manassas, VA) supplemented with 10% charcoal-stripped FBS (Sigma Aldrich), 0.5% penicillin/streptomycin, 2 mM L-glutamine (Sigma Aldrich), 100 μg/mL gentamicin (G18; Mediatech Inc.), 1 μg/mL hygromycin B (Life Technologies, Carlsbad, CA), and 5 μg/mL blasticidin (Invitrogen, CA). Cells were grown at 37°C and 5% CO₂ until reaching 80% confluency.

Primary decidual cells were isolated using the standard isolation method as previously described [25, 33]. After passage 3 or 4, decidual cells (~1X10 [5]) were immortalized using retroviral transduction supernatant obtained from PA317 LXSN 16E6E7 cells (ATCC CRL-2203) through inducing Human Papillomavirus E6 and E7 (HPV E6/E7) oncogene [34]. The supernatant also contained 20 μL of 1 g/μL protamine sulfate and 1800 μL serum-free DMEM/F12 medium. Transduced cells were selected in the presence of 50 μg/mL G418 (Mediatech Inc.). Established cell lines were kept routinely in T75 flasks containing regular media composed of DMEM/F12 medium supplemented with 5% heat-inactivated FBS (Sigma Aldrich), 1% amphotericin B, and 1% penicillin/streptomycin (Mediatech Inc.) at 37°C and 5% CO₂ until reaching 70–80% confluence.

Immortalized human endocervical epithelial cells (EECs) from hysterectomy material from women with benign gynecological conditions were used in this study [34]. EECs were cultured in keratinocyte serum-free medium, a culture medium highly selective for epithelial cells, supplemented with bovine pituitary extract (30 μg/mL), epidermal growth factor (0.1 ng/mL), CaCl₂ (0.4 mM), and primocin (0.5 mg/mL) (ant-pm-1; Invitrogen, San Diego, CA) at 37°C and 5% CO₂; cells were grown to 80% confluence.

Studying the uptake of AEC-derived exosomes and MVs by maternal cells, immunocytochemistry staining

Myometrial, decidual, and cervical cells were each cultured on an eight-well slide (~50 000 cells/well) for 24 h. After 24 h, the culture media were removed, cells were washed with PBS, and cells were treated with media containing DiO-labeled control exosomes (n = 3) and FITC-conjugated Annexin-V-labeled control MVs (n = 3). Negative controls (n = 3) of each cell line were cultured without the addition of exosomes/MVs. After a 4 h incubation at 37°C and 5% CO₂, media were removed, cells were washed with PBS, fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X, and blocked with 3% BSA in PBS for 1 h. Cells were incubated with primary antibodies to alpha-smooth muscle actin (1A4), eBioscience (Catalog# 14-9760-82, Thermo Fisher Scientific) for myometrial and decidual, or cytokeratin-18 (1 μL/mL, Catalog# ab668, Abcam) and vimentin (3.7 μL/mL, Catalog# ab92547, Abcam) for cervical cells, in 1:500 dilution in 3% BSA/PBS, rocking overnight at 4°C. Slides were washed three times with PBS and incubated with secondary antibody Alexa Fluor 594 or 647 (Life Technologies) diluted 1:400 in PBS for 1 h in the dark at room temperature. Slides were washed with PBS, 4,6-diamidino-2-phenylindole (DAPI; Invitrogen, Thermo Scientific) was added, and then slides were washed again and mounted using MOWIOL 4-88 (Sigma Aldrich) mounting medium. Slides were dried overnight and then the cells were imaged using a BZ-X810 fluorescence microscope (Keyence BZ series, USA). Images were obtained and analyzed using Image J (National Institutes of Health; rsbweb.nih.gov/ij; version 1.51) to confirm the uptake of EVs by the cells. Three-dimensional reconstructions of the cells were also created. Image analysis was conducted in triplicate for all cell experiments.

Studying the functional effect of AEC-derived exosomes and MVs in recipient cells

Myometrial, decidual, and cervical cells (n = 8 for each cell line) were cultured in 48-well plates and incubated overnight. After 24 h, cells were washed with PBS, and media were replaced with exosome-free cell culture media. Cells were treated with 1 × 10⁷ particles/well of control and CSE exosomes and MVs for 24 h. Negative and positive control wells (n = 3) were included that consisted of exosome-free media only and Tumour Necrosis Factor alpha (TNF-alpha) (50 ng/mL; Sigma Aldrich), respectively. After the completion of the treatment, the media samples were collected from each well and stored at −80°C.

As a control experiment, the action of exosomes and MVs in maternal cells was blocked to determine whether the observed effects were mediated by EVs. This was done through the incubation of decidual, myometrial, and cervical cells with exosomes and MVs at 4°C for 6 h. The media were then collected as described above and stored at −80°C until analysis.

NF-kappaB translocation and activation

Maternal cells treated with exosomes and MVs or control cells were fixed and premetallized after media collection as described above. Cells were incubated with anti-NF-kappaB p65 antibody (pS529; BD BioSciences), 1:100 dilution in 3% BSA in PBS, for 24 h at 4°C. On the second day, cells were washed and incubated in the dark with the secondary antibody (Alexa Fluor 594) diluted at 1:400 in PBS. After washing, nuclei were stained with DAPI, and then the cells were imaged using the BZ-X810 fluorescence microscope (Keyence BZ series, USA).

Multiplex cytokines measurement by Luminex

To examine whether exosomes and MVs can elicit an inflammatory response in maternal uterine tissues, media collected from maternal cells after treatment with control and CSE-treated AEC-derived exosomes and MVs were quantitated for inflammatory markers (i.e., granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin 6 (IL-6), and IL-8.) using a Milliplex MAP panel and detection kit (EMD Millipore Corporation, Billerica, MA, USA). First, the collected media were thawed and spun to remove any cell debris. Media and substrates were added to the kit plate and instructions were followed according to the kit protocol.

Briefly, the plate was washed with wash buffer by shaking for 10 min and then decanted. Standards, controls, samples, assay buffer, and magnetic beads were added to the plate. Media for each cell culture were used as background controls. The plate was sealed and incubated with agitation on a plate shaker overnight (16–18 h) at 4°C in the dark. The plate contents were gently removed and washed three times with wash buffer. 1X Milliplex MAP detection antibody was added to the plate and incubated on a shaker for 1 h at room temperature (20–25°C) in the dark; 1X streptavidin–phycoerythrin was added and incubated on a plate shaker for 30 min in the dark. After aspirating the plate content and washing for three times, an amplification buffer was then added. The plate was run on a Luminex 200 (LX200-XPON-IVD, Luminex Corporation, Austin, TX, USA) apparatus. Collected data were analyzed.

Statistical analysis

The distributions of the three inflammatory markers, namely GM-CSF, IL-6, and IL-8, were examined, and mean and standard deviations were calculated for each paired group (e.g., decidual control MVs vs decidual CSE-MVs). To determine whether analyte levels significantly differed between conditions for each cell type, a paired T-test was conducted (P < 0.05). The false discovery rate was used to correct for Type I error. All analyses were conducted using SAS V9.2 (Cary, NC).

Results

Isolation and characterization of AEC-derived exosomes and MVs

Exosomes and MVs were isolated from AEC cell culture media by the differential centrifugation method (Figure 1). The size and concentration were determined by Nanosight N300, which showed that MVs had an average size range of between 160 and 400 nm and a concentration of 3.0 × 10⁹ particle/mL, whereas exosomes had an average size of 90–150 nm and a concentration of 2.0 × 10¹⁰ particles/mL (Figure 2A and B). The shape and structure of isolated exosomes and MVs were studied by TEM, confirming big-size circular vesicles referred to as MVs and smaller-size cup-shaped vesicles referred to as exosomes (Figure 2C and D). Although exosomes are also circular, they appear as cup-shaped under TEM, confirming two distinct populations in our preparations.

Figure 2.

Characterization of isolated exosomes and MVs by Nanosight, transmission electron microscopy (TEM), dot-blot, western blot, and immunofluorescence staining: (A) concentration (particles/mL); (B) size (nm) of exosomes and MVs measured by Nanosight (n = 8); (C) TEM showing the morphology and size of MVs; (D) TEM showing the morphology and size of exosomes. Arrows indicates extracellular vesicles; (E) dot-bot for MVs fraction showing the expression of FLOT1, ICAM, CD81, CD63, EpCAM, ANXA5, and TSG101; (F) dot-blot for exosomes showing positive stain for FLOT1, ICAM, ALIX, CD81, CD63, EpCAM, ANXA5, and TSG101; (G) western blot showing OCT-4 and HLA-G proteins were expressed on MVs, whereas HSP70 and NANOG were expressed on exosomes. Exosome markers CD81 and CD63 were highly expressed on exosomes than MVs. Images are cropped to improve quality; (H) Annexin-V-FITC staining for exosomes and MVs. Left: FITC-conjugated Annexin-V-labeled MVs, scale bar = 500 μm; middle: FITC-conjugated Annexin-V-labeled exosomes, scale bar = 500 μm; right: negative control, scale bar = 200 μm. Annexin-V-FITC staining was used to differentiate between exosomes and MVs. Annexin-V binds to phosphatidylserine (PS), which is exposed to the outer leaflet of MVs, but not exosomes, during biogenesis.

Exosome antibody array (dot-blot) and western blot

Marker characterization using dot-blots revealed the presence of the typical exosomal marker proteins CD81, CD63, Annexin A5 (ANXA5), Intercellular Adhesion Molecule 1 (ICAM), Flotillin1 (FLOT1), ALIX, Epithelial Cellular Adhesion Molecule (EpCAM), and Tumor Susceptibility 101 (TSG101) in exosomes (Figure 2E). Interestingly, the same markers were observed in MVs except for ALIX, which was negative (Figure 2F). GM130 was negative in both exosomes and MVs, excluding any putative contaminants such as cell debris.

To identify some specific markers for AEC-derived MVs that could distinguish them from exosomes, western blots for OCT-4, NANOG, HLA-G, HSP70, CD63, and CD81 were performed. OCT-4 and HLA-G were exclusively expressed on MVs, whereas HSP70 and NANOG were expressed on exosomes only (Figure 2G). We have already reported the absence of OCT-4 in AEC-derived exosomes in our prior reports [26] and were able to replicate those data. Although not quantitative, comparing the intensity of the exosomal markers CD81 and CD63 between exosomes and MVs suggested that these markers were highly expressed on exosomes, as expected, than to MVs.

Staining exosomes and MVs with FITC-conjugated Annexin-V

In order to differentiate between exosomes and MVs on a molecular level, both were stained with FITC-conjugated Annexin-V. Annexin-V interacts with phosphatidylserine (PS) that is commonly expressed on the outer leaflets of MVs during biogenesis. MVs showed Annexin-V staining, whereas exosomes did not show any green fluorescence staining, confirming the purity of the isolated EVs subtypes (Figure 2H).

Proteomics analysis of AEC-derived EVs

The protein cargo for AEC-derived exosomes and MVs under normal and OS conditions was determined by LC/MS. Mass spectroscopy identified a total of 1263 proteins in control exosomes, 1282 in CSE-exosomes, 982 proteins in control MVs, and 885 proteins in CSE-MVs. The Venn diagram in Figure 3A (Supplementary Table S1) shows the common and unique proteins between samples. It was observed that the OS condition resulted in changes in the cargo of exosomes and MVs compared with control. The common proteins identified among all samples were Flotillin-10 (FLUT10), Annexin (A1, A2, A3, A4, A5, A6), CD63, coagulation factor V, mitogen-activated protein kinase (1,2,3), EH Domain Containing 3 (EHD3), ESCRT complex proteins such as TSG101, Charged multivesicular body protein 4a (CHMP4A), and Vacuolar Protein Sorting 13 Homolog D (VPS13D), NEDD4 E3 ligase (ITCH), fibroblast growth factor 4 (FGF4), Major Histocompatibility Complex, Class I, H (HLA-H), several proteasome subunits, and Vesicle Associated Membrane Protein 7 (VAMP7). Examples of proteins that were specific to control exosomes when compared with CSE-exosomes were Heat Shock Protein Family D (Hsp60) Member 1 (HSPD1) (apoptosis-related network), HSP105, prostaglandin reductase 3, histone deacetylase (HDAC3), calpain 2 subunit, integrin-beta-6, serine/threonine protein phosphatase 2A (PPP2CB) function for TGF and Wnt signaling pathways, BAG4 (oxidative damage and TNF signaling), IL-10, ARF-GAP, and BCLAF1. On the other hand, proteins such as calcium/calmodulin-dependent protein kinase, serine/threonine protein kinase (mTOR), IL-15, Janus kinase 3 (JAK3), voltage-dependent T-type calcium channel subunit alpha, complement factor I, Mitogen-activated protein kinase (MAPK), and Matrix Metallopeptidase 9 (MMP9) were exclusive to CSE-exosomes compared with control exosomes. MVs showed expression of 182 specific proteins under control condition when compared with CSE such as calnexin, BAG4 (TNF signaling), fibrinogen beta, HLA, MAP3K12-binding inhibitory protein, integrin-beta-4, Wnt 5A, and programmed cell death 6 interacting protein (function for endocytosis and budding of viruses and vesicles). Comparing CSE-MVs with control MVs showed the expression of 85 unique proteins, such as E3 ubiquitin protein ligase (RNF123 and RNF139), HSP70, GTPase activating protein, cell adhesion molecule, mitogen protein kinase (MAPK1 and MAPK3K10), prostaglandin E synthase, voltage-dependent T-type calcium channel subunit alpha, AKAP6 (G protein signaling), angiopoietin 4 (Ras and p13-serine/threonine kinases (AKT) signaling), BAG1 (regulation of HSF-1-mediated heat shock response), Wnt4, and RIF1 (telomerase association protein). Several Ras-related proteins were differentially expressed between the four samples. Ras-GTPase activation was observed in both exosomes and MVs under OS conditions.

Figure 3.

Qualitative and quantitative proteomic analyses of AEC-derived exosomes and MVs fractions: (A) Venn diagram showing the distribution of identified proteins in each fraction. Each Venn diagram represents the unique and common protein between control exosomes and control MVs; control exosomes and exosome under OS conditions (CSE-Exo), control MVs and MVs under OS conditions (CSE-MVs); and exosomes and MVs under OS conditions (CSE-Exo and CSE-MVs, respectively); (B) volcano plot showing upregulated and downregulated proteins for CSE-Exo vs Exo-control, CSE-MVs vs control MVs, control Exo vs control MVs, and CSE-Exo vs CSE-MVs. Significant proteins (adjusted P < 0.05) with log2-fold change ±0.6 are represented by green dots. Nonstatistically significant differential proteins (P > 0.05) are represented by gray dots; (C) clustered heatmap for the relative abundance ratio of detected proteins in control MVs vs control exosomes, CSE-MVs vs control MVs, CSE-Exo vs control Exo, and CSE-MVs vs CSE-Exo. Proteins are expressed on a log2-fold change and color coded. Blue represents downregulation, red represents upregulation, and white represents no change.

Some of the exosomal-specific markers, such as CD63, FLOT1, and TSG101, were expressed in all samples; however, CD81 was absent in CSE-MVs, but was expressed in exosomes (control and CSE) and control MVs. Interestingly, CD9 was only expressed in exosomes samples (control and CSE), indicating the purity of the isolation method for the EVs subpopulation. In addition, HSP70 was absent in control MVs, but was expressed in exosomes (control and CSE) and CSE MVs. Some of the specific proteins that can be used to differentiate MVs from exosomes (under control condition) included calnexin, FGF2, NOTCH1, ESCRT complex protein VPS37D, mTOR, PTEN, CAMKK2, LAMB2, PLB1, clathrin light chain (CLTA), and Wnt 5B. Although HLA-G, OCT-4, and Nanog were detected by western blot, they were absent in the proteomics analysis.

Subsequently, statistical analysis between each pair of groups was carried out and presented in the form of a volcano plot (Figure 3B) to show the upregulated and downregulated proteins in terms of a log2-fold change ±0.6 (P < 0.05). Rab1A, Rab8A, complement factor 1, VAMP3, Ras-related GTP-binding protein A, syntain1A, Annexin A11, and Eukaryotic Translation Initiation Factor 2 Subunit Alpha (ELF-2alpha) kinase were significantly downregulated in CSE exosomes vs control, whereas Ras GTPase-activating protein, Rab11A, Rab11B, Rab 19, Rab23, Rab39A, cell adhesion molecule 2, coronin 1A, histone 2A, Annexin A, arrestin domain-containing protein 1 (ARRDC1), HSP70, HSP105, integrin-beta6, voltage-dependent T-type calcium channel alpha subunit, and Mitogen-activated protein kinase (MAPK) kinase were significantly upregulated. The proteins that were significantly upregulated in CSE-MVs vs control included ARRDC1, clathrin light chain, complement factor B, complement factor 1, HSP70, integrin-beta6, MAPK kinase, RAB 11B, RAB 22A, and voltage-dependent T-type calcium channel alpha subunit, among others, whereas RAB 1A, 3A, calnexin, annexin A11, and CD63 were significantly downregulated.

The differential expression of all proteins identified in the EV subpopulation under normal and OS conditions was presented in the form of a clustered heatmap showing fold change for each pair of groups (Figure 3C).

IPA

A heatmap for canonical pathways controlled by proteomic cargo in exosomes and MVs treated with CSE compared with control is presented in Supplementary Figure S1. The upregulated and downregulated pathways are shown in red and green, respectively. NRF-2-mediated OS response was upregulated in CSE-exosomes and CSE-MVs when compared with control, confirming our recent reports on OS-induced NRF-2 activation in human fetal membranes [35]. Other pathways that were upregulated in both CSE-exosomes and CSE-MVs compared with control were acute phase response signaling, NF-kappaB, regulation of epithelial–mesenchymal transition by growth factor, IL-8, Mitogen-Activated Protein Kinase (ERK5), Janus kinase (JAK)/signal transducer and activator of transcription STAT, natural killer cell, IL-6, nuclear factor of activated T cells (NFAT) (regulation of immune response), GM-CSF, and senescence pathways. Pathways that were exclusively upregulated in CSE-exosomes compared with control include GP6 signaling, IL-15, Liver-X-receptor/retinoid X receptor (LXR/RXR) activation, integrin signaling, CD28 signaling in T helper cells, coagulation system, and ERK/MAPK signaling. CSE-MVs showed upregulation of TGF-beta and STAT3. MVs were also compared with exosomes under normal and OS conditions, which showed the downregulation of most detected pathways except for peroxisome proliferator-activated receptors/retinoid X receptor (PPAR alpha/RXR apha), nitric oxide synthase 3 (eNOS), and apoptosis signaling. However, under OS condition, CSE-MVs showed upregulated NRF2, P13K, endolthelin1, nitric oxide, and reactive oxygen species (ROS) production in macrophages, JAK/STAT, natural killer cells (NK), IL-6, coagulation system, and ERK/MAPK when compared with CSE-exosomes.

This indicates that the cargo in exosomes and MVs has some similarities and yet enough differences that may contribute to differential response in recipient cells.

Studying the uptake of AEC-derived exosomes and MVs by maternal cells, immunocytochemistry staining

The localization of exosomes and MVs in recipient decidual, myometrial, and cervical cells was visualized by fluorescence microscopy and Z-stack analysis (Figures 4–6). FITC-conjugated Annexin-V-labeled MVs (Figure 4) and DiO-labeled exosomes (Figure 5) were detected in recipient maternal cells, indicating their uptake by maternal cells. Figure 6 shows Z-stack analysis and three-dimensional reconstructions that confirmed the location of EVs within the maternal cells.

Figure 4.

Uptake of Annexin-V-FITC-labeled MVs by maternal cells. Left: Decidual cells, middle: myometrial cells, and right: cervical cells. (A) Cell-specific marker: alpha-smooth muscle actin (for decidual and myometrial cells), vimentin (for cervical cells). (B) Nuclei staining by 4′,6-Diamidino-2-phenylindole (DAPI); (C) Annexin-V-FITC-labeled MVs; (D) merged images. Scale bar = 100 μm. Arrows indicate the location of MVs in close-up view.

Figure 6.

(A) Colocalization of Annexin-V-labeled MVs and (B) DiO-labeled exosomes in maternal cells. Left: ImageJ 3D viewer for maternal cells showing colocalization of MVs and exosomes, middle: close-up image for the colocalization of MVs and exosomes inside maternal cells, right: colocalization analysis showing MVs or exosomes (green) overlapping within maternal cells (red) and nuclei (blue). Cell-specific marker: alpha-smooth muscle actin (for decidual and myometrial cells), cytokeratin (for cervical cells).

Figure 5.

Uptake of DiO-labeled exosomes by maternal cells. Left: Decidual cells, middle: myometrial cells, and right: cervical cells. (A) Cell-specific marker: alpha-smooth muscle actin (for decidual and myometrial cells), vimentin (for cervical cells). (B) Nuclei staining by 4′6-Diamidino-2-phenylindole (DAPI); (C) DiO-labeled exosomes; (D) merged images. Scale bar = 100 μm, scale bar for cervical cells = 200 μm. Arrows indicate the location of MVs in close-up view.

Activation of NF-kappaB

After the incubation of maternal cells with AEC-derived exosomes and MVs for 24 h, cells were stained using anti-NF-kappaB p65 antibody to determine the activation (phosphorylated NF-kappaB) and nuclear translocation of NF-kappaB. The nuclei were stained by DAPI for localization. AEC-derived EVs under control and OS conditions activated NF-kappaB, which translocated to the nucleus, shown as red fluorescence, in the decidual, myometrial, and cervical cells (Supplementary Figure S2). TNF-alpha, a known activator of NF-kappaB instead of EVs, was used as a positive control, which also showed activation of NF-kappaB in a similar pattern compared with AEC-derived exosomes and MVs.

Functional role of AEC-derived exosomes and MVs in maternal cells

To study whether AEC-derived EVs have a pro-inflammatory property in maternal cells, the concentration of pro-inflammatory cytokines (GM-CSF, IL-6, and IL-8) was measured in the decidual, myometrial, and cervical cell culture supernatant after incubation with AEC-derived exosomes and MVs (either from control AECs or CSE-treated AECs) for 24 h (Figure 7). The concentration of cytokines was also measured in untreated cells (negative control) and cells treated with TNF-alpha (positive control).

Figure 7.

Cytokine concentrations in maternal cells after incubation with AEC-derived control and CSE-EVs: (A) decidual cells, (B) myometrial cells, (C) cervical cells. Control (negative control): TNF-alpha (positive control; cold) cells were incubated at 4°C to prevent EV endocytosis being used. The concentrations of GM-CSF, IL-6, and IL-8 were measured. Values represent mean ± SD (n = 8). ^*P < 0.05; ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001.

Treating decidual cells with CSE-exosomes resulted in significantly higher concentrations of GM-CSF (P < 0.01), IL-6 (P < 0.001), and IL-8 (P < 0.05) in comparison with control exosomes (Figure 7A, Supplementary Table S2). In addition, the concentrations of GM-CSF and IL-6 were significantly higher in decidual cells when treated with CSE-MVs compared with control MVs (P < 0.001; P < 0.001, respectively). However, IL-8 concentration was not statistically different in decidual cells treated with CSE-MVs (P > 0.05).

In myometrial cells (Figure 7B, Supplementary Table S3), the concentration of IL-8 was significantly higher after incubating the cells with CSE-exosomes compared with control exosomes (P < 0.05). In contrast, the concentration of GM-CSF in myometrial cells was significantly lower after incubation with CSE-exosomes treatment (P < 0.01). Interestingly, the concentrations of GM-CSF and IL-6 in myometrial cells were significantly higher after treatment with CSE-MVs compared with control MVs (P < 0.05; P < 0.01, respectively).

In cervical cells (Figure 7C, Supplementary Table S4), there were no significant differences in the levels of cytokines except for GM-CSF, which was only significantly higher after incubating the cells with CSE-MVs compared with control MVs (P < 0.05).

All maternal cells treated with TNF-alpha (positive control) showed higher cytokine levels than untreated cells. As expected, the concentration of cytokines remained very low in all cold controls. Exosomes uptake by recipient cells commonly occurs through energy-dependent endocytosis [26]. The inhibition of cytokine production after incubation of maternal cells in a cold temperature (4°C) treated with exosomes and MVs indicates that the observed effect was related only to exosomes and MVs. Also, this confirms that the uptake of exosomes and MVs occurred predominantly through endocytosis.

Discussion

A substantial body of evidence exists regarding the role of exosomes in pregnancy and labor. Specifically, senescent fetal cell-derived exosomes at term induce inflammatory response on the maternal side to contribute to parturition [20, 21, 25, 26, 35]. There is a gap in the literature about the specific properties and functional role of MVs, which are often confused with exosomes [21]. In this report, we tested the hypothesis that MVs, as a paracrine communicator between the fetus and the mother, could provide a complementary functional role that supports the role of exosomes. The work presented herein provides (i) successful isolation and characterization of exosomes and MVs from AECs; (ii) differences between exosome and MV cargo contents as well as some specific markers to distinguish two populations of EVs; and (iii) distinct functional roles of AEC-derived exosomes and MVs in maternal cells under normal and OS conditions and their impact on the level of pro-inflammatory cytokines.

The first challenge in studying EVs is their isolation. Since there are overlapping properties, such as size and markers between MVs and exosomes [36], the isolation of a pure fraction of MVs and exosomes from a heterogenous pool of EVs mandates stringent and specific optimization approaches. Several methods have been reported in the literature, including size exclusion, immunoaffinity, ultrafiltration, commercial kits, and differential centrifugation [37]. The differential centrifugation method is considered the most popular method used for the isolation of different EV subpopulations. It provides several advantages, such as a high yield, flexibility in modifying any step in terms of centrifugation speed/time, and the incorporation of an additional filtration/washing step to meet the optimum criteria of the required product, and it does not require the use of specific antibodies as in the immunoaffinity method, which avoids bias in the composition of isolated EVs [38]. In this study, we used an optimized differential centrifugation method by applying and testing the outcome of several centrifugation and filtration steps to ensure successful segregation of MVs and exosomes particles into two clear fractions whose characters were further confirmed using multiple approaches. The biogenesis of MVs includes the externalization and exposure of PS to the outer leaflet of the vesicle membrane, which can be considered a key difference between MVs and exosomes [39]. In contrast, PS remains inside the membrane of exosomes during biogenesis. These data provided validation for our methodological approaches and the purity of our preparations prior to testing their functional properties.

Interestingly, we found that HLA-G and OCT-4 were present only in MVs. AECs express surface HLA-G and OCT-4 [40, 41]. Since the biogenesis of MVs includes budding and fission from the surface of cell membrane, it is not surprising that MVs will carry some of their parent cell surface proteins as part of their membrane. Furthermore, HLA-G plays a crucial role in immunomodulation during pregnancy [42]; thus, MVs can act as an important shuttle for HLA-G and other protein molecules from fetal cells to other cells. The lack of HLA-G on exosomes further suggests their functional differences in regulating immune responses at the feto-maternal interfaces.

Furthermore, we identified NOTCH1 protein in control MVs only. In addition, TSG101, NEDD4 E3 ligases (ITCH), and multiple components of the ESCRT complexes (CHMP and VPS proteins) were identified in exosomes and MVs under normal and OS conditions. These proteins are required in recruiting NOTCH receptors from the plasma membrane into EVs [43]. NOTCH receptors are specific plasma membrane proteins. They have several physiological functions, such as in trophoblast function, placental angiogenesis, tissue homeostasis, immunomodulation, and embryonic development [43, 44]. NOTCH1 has also been shown to play a role in inflammation-induced preterm birth [45]. NOTCH signaling is mediated through the canonical pathway, which requires cell-to-cell direct contact, or the noncanonical one, which is ligand independent. A previous study by Wang and Lu [43] identified NOTCH receptor in special-type MVs identified as ARRDC1-mediated MVs. The study suggested that these types of MVs are capable of recruiting NOTCH receptors and mediating noncanonical NOTCH signaling. Our finding suggests that AEC-derived MVs under normal conditions might be capable of performing a similar function and of mediating noncanonical NOTCH signaling. Interestingly, NOTCH1 was not found in CSE-MVs like exosomes, suggesting that exposing AEC to OS conditions affects the functional expression of NOTCH1 on MVs as well as activating the NOTCH signaling pathway. This indicates that NOTCH1 is carried only through MVs and under noninflammatory conditions. Another protein marker that was found to be specific to MVs was calnexin. Previous studies have shown that calnexin could be specific to some MVs [38].

Parturition is considered an inflammatory state that is characterized by the elevation of inflammatory cytokines and chemokines and leukocyte activation in the myometrium, decidua, and cervix as well as peripheral blood [46]. This results in increasing myometrial contractility, activation of decidual/fetal membranes, and cervical ripening [47]. Fetal inflammatory paracrine signaling via EVs can elicit changes in maternal cells to coordinate the process of parturition. We have previously demonstrated that that compounds in CSE induces amnion cell senescence via a mechanism involving ROS and DNA damage [48]. Both pathways may contribute to preterm birth and Preterm premature rupture of membrane (pPROM). Thus, we used CSE to induce sterile inflammation and simulate OS state in AECs.

Several differentially expressed proteins were identified between AEC-derived exosomes and MVs under normal and OS conditions. Most of these proteins reflect a state of inflammation and compromised immunoregulation responses, which were expected to be higher under OS conditions than control. Moreover, Ras-related proteins were differentially expressed among exosomes and MVs under normal and OS conditions. For example, Ras-GTPase activating proteins were upregulated in exosomes and MVs under OS conditions compared with control. Studies have shown that Ras-GTPase has been linked to inflammation and preterm birth through OS [49, 50]. Thus, exosomes and MVs reflect a real-time gestational state. In this report, we recreated contributions of human-term fetal membranes to a normal-term labor condition by exposing AECs to OS conditions as expected in utero. OS at term accelerates senescence and produces exosomes and MVs enriched in inflammatory cargo. We showed here that both exosomes and MVs can cause paracrine signaling and elevate the production of inflammatory cytokines (IL-6, IL-8, and GM-CSF) that are associated with labor-associated inflammatory changes in maternal tissues [29, 51]. However, cervical GM-CSF response was lower in response to both OS-induced AEC-derived exosomes and MVs compared with respective controls. The extent of cytokine response by various cell types to either of the vesicles is different, and specific cytokine responses are cell type specific. We have examined only a limited number of inflammatory mediators and therefore the discussion is limited to an overall pro-inflammatory change in maternal uterine cells. In addition, the preliminary study carried to detect the activation of NF-kappaB in maternal cells after incubation with exosomes and MVs provides an interesting finding for NF-kappaB activation and translocation into the nucleus under normal and OS conditions. However, since the study was performed at only one time point (after 24 h), no quantification for NF-kappaB was performed. Thus, we cannot compare the level of NF-kappaB activation between different cells. Further studies are required to assess the effect of exosomes and MVs on NF-kappaB activation.

In fact, the demonstrated changes in the uterine cavity are governed by multiple complex factors; thus, in vivo models that can provide a synergistic effect between different uterine tissues and paracrine, endocrine, and immune mediators could be used to further study the effect of exosomes and MVs during pregnancy. Regardless, data demonstrate that fetal cell-derived particles carrying pro-inflammatory cargo can elicit an inflammatory reaction in maternal tissues.

We have not examined contributions by specific vesicular cargo on maternal cells. For example, the increase in calcium influx into myocytes is associated with labor initiation, which leads to an increase in myometrium contractility [52, 53]. Interestingly, exosomes and MVs under OS conditions showed an increase in the expression of the voltage-dependent T-type calcium channel alpha subunit. This finding suggests that exosomes and MVs could play a role in increasing myometrium contractility through increasing the expression of calcium receptors and calcium influx. Future studies may examine the contributions of individual cargo to given cell types.

It is not surprising that even exosomes and MVs under control condition showed some inflammatory response in uterine cells. Studying the cargo of exosomes and MVs through proteomics analysis has revealed the presence of proteins mediating different inflammatory signaling such as cell adhesion molecule, Ras-related protein, p53 signaling, MAPK signaling, and TNF signaling. Therefore, it can be concluded that functional differences between AEC-derived exosomes and MVs in uterine cells were subtle as both could induce a similar pattern of inflammatory changes. Further studies on other cytokines/chemokines are needed to further assess the functional differences between exosomes and MVs.

Although the current study has demonstrated the effect of fetal-derived exosomes and MVs on maternal cells, a recent study showed that fetal-derived EVs could not be detected in uterine tissue using an animal model [54]. The lack of in vivo model to study the actual effect of exosomes and MVs on uterine tissues is considered a limitation of our current study. Identification of specific markers for placental and fetal-derived EVs in vivo will provide a better understanding of their specific roles. In addition, further studies are required to assess the influence of blocking one type of EVs on the other and to study the impact of this on cell survival. This would provide a comprehensive insight into the functional role of EVs [55].

Conclusion

In summary, this study highlighted the differences between exosomes and MVs derived from AECs regarding their cargo and functional properties in feto-maternal communication. We have successfully established an optimized differential centrifugation method that can be applied to isolate different EVs subpopulation. The proteomics analysis identified a set of unique protein markers that can be used to differentiate between AEC-derived exosomes and MVs. However, the lack of universal specific markers for MVs is still a challenge that needs further investigation. We have also shown that not only exosomes but also MVs can carry pro-inflammatory signals and can traffic between fetal and maternal tissues. Exosomes and MVs might be different in their biogenesis, cargo, and properties, but they share a similar functional role in uterine cells. Exosomes and MVs together form a sophisticated communication network between fetal and maternal tissues for signal propagation during parturition. Exosomes and MVs provide insight into several cellular changes and thus can be developed as biomarkers that reflect the underlying physiological state of gestational tissues. Furthermore, MVs can be used as an approach to deliver large drug compounds that cannot be efficiently entrapped inside the smaller-sized exosomes.

Grant Support: This study is supported by grant funds from National Institutes of Health (NIH)/National Institute of Allergy and Infectious Diseases (NIAID) (R21AI140249 to RM).

Data Availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Authors’ Contributions

HIS contributed towards the investigation, methodology, formal analysis, data curation, writing, review, and editing; ER contributed into the methodology (cell culture); OAGT contributed to the methodology and formal analysis; TK conducted the Multiplex Luminex Assay; AKK contributed to the methodology (proteomics study); SS-M contributed to the investigation, methodology, formal analysis, and data curation; BDT contributed to the statistical analysis and reviewing and editing; RM contributed to the supervision, conceptualization, funding, writing, and reviewing and editing.