Current Topics: Reviews
Transport Characteristics of Placenta-Derived Extracellular Vesicles and Their Relevance to Placenta-to-Maternal Tissue Communication
2022 Volume 70 Issue 5 Pages 324-329
Details
2022 Volume 70 Issue 5 Pages 324-329
The placenta, a unique organ that helps maintain a healthy pregnancy, plays a pivotal role in maternal adaptation to pregnancy and releases extracellular vesicles (EVs), autacoids, and hormones. EVs are membranous vesicles released by all types of cells, including placental trophoblasts, which are involved in intracellular communication by delivering their cargo, such as proteins, nucleic acids, and lipids, to the targeted cells in a neighboring or distant location. Recently, an increasing number of publications have reported that EVs secreted from the placenta into maternal circulation deliver their cargo to maternal organs and mediate placenta-to-maternal communication during pregnancy. This review provides an overview of the transport mechanism of placenta-derived EVs to maternal organs.
The placenta is an organ that helps maintain a healthy pregnancy. It contains trophoblasts that, in humans, are histologically organized into extravillous trophoblasts, cytotrophoblasts, and syncytiotrophoblasts. The placenta provides nutrient supply and gas exchange, waste removal, and immune support functions for the fetus, and it mediates molecular exchanges between the maternal and fetal systems.1) The placenta plays a pivotal role in orchestrating the maternal adaptation to pregnancy and releases many different molecules that can alter the maternal physiology.2) During pregnancy, the placenta, particularly the syncytiotrophoblasts, secretes a large range of extracellular vesicles (EVs) of different sizes, such as macrovesicles (20–100 µm), microvesicles (100–1000 nm), and nanovesicles (30–150 nm), including exosomes.3,4)
EVs are released into the maternal and fetal circulation during pregnancy and contain various types of cargo, such as proteins, nucleic acids, and lipids.4–8) The cargo carried by EVs can be transported to target cells in neighboring or distant locations, thereby modulating the maternal physiology and fetal development.9,10) Thus, EVs derived from placental trophoblasts can communicate with any tissue in the fetus and mother. As gestation progresses, the concentration of EVs, especially exosomes derived from the placenta, increases in the maternal plasma,5,6) and their bioactivity may change under conditions of disease, including preeclampsia and gestational diabetes.11–13) Therefore, in recent years, the utility of EVs for the early diagnosis of pregnancy complications has been well documented.14–17) In this review, we summarize the current knowledge regarding the transport mechanisms of placenta-derived EVs in the placenta and in the maternal organs. We believe that clarifying the EV-based mechanisms underlying delivery would lead to not only a further understanding of placenta-derived EV distribution and physiological changes in the mother during pregnancy but also the potential usefulness of the development of drug delivery of large molecules, such as peptides and nucleic acids.
The uptake of EVs, especially exosomes, by target cells involves various mechanisms, including endocytosis, macropinocytosis, phagocytosis, lipid raft-mediated internalization, and membrane fusion.18) EV uptake is largely dependent on the surface molecules and glycoproteins on the membrane of the EVs and target cells.18,19) Recently, several proteomic studies on placenta-derived EVs isolated from the blood of pregnant women, placental explants, or trophoblastic cell lines have been reported.3–5,20–22) Although the composition of EVs is dependent on the tissue or cell type of origin, EVs commonly express specific proteins, such as endosomal protein markers (CD9, CD81, and CD63) and cell adhesion molecules (integrins and intercellular adhesion molecules), on their surface. In contrast, placenta-derived EVs are unique compared to other EVs in terms of their expression of placenta-specific proteins, including placental alkaline phosphatase (PLAP), syncytin-1, syncytin-2,4,23–27) CD276,4) and human leukocyte antigen G (HLA-G).25) These proteins on the surface of placenta-derived EVs have been reported to mediate their internalization by maternal organs during pregnancy (Fig. 1).
Placenta-derived EVs are transported to the maternal interstitial lung macrophages, liver endothelial cells, and liver Kupffer cells in an integrin-dependent manner. Placenta-derived EVs are transported back to placental syncytiotrophoblasts, which are mediated by syncytin-1-alanine-serine-cysteine transporter 2 (ASCT2) and syncytin-2-major facilitator superfamily domain containing 2A (MFSD2A) interactions. Placenta-derived EVs can also interact with vascular endothelial and immune cells.
Syncytiotrophoblasts form the placental barrier, which is located between the maternal and fetal blood. Syncytiotrophoblasts play a pivotal role in exchanging nutrients and gases, as well as in synthesizing and secreting autacoids, hormones, and EVs.1,28) The syncytiotrophoblastic layer is formed from the fusion of cytotrophoblasts. This process is mediated by syncytin-1 and syncytin-2, which are encoded by the human endogenous retrovirus (HERV) family of genes.29–32) Syncytin-1 and syncytin-2 retain the fusogenic ability of the retroviral envelope when they bind to their receptors, alanine-serine-cysteine transporter 2 (ASCT2) and major facilitator superfamily domain containing 2A (MFSD2A), respectively. This causes fusion of the syncytiotrophoblasts and, consequently, leads to the formation of the multinuclear syncytiotrophoblastic layer. Vargas et al. demonstrated that both syncytin-1 and syncytin-2 are highly expressed on the surface of exosomes derived from syncytiotrophoblasts and the BeWo human choriocarcinoma cell line. These exosomes are rapidly taken up by BeWo cells in a syncytin-1- and syncytin-2-dependent manner.23) The immunofluorescence analysis of exosome distribution demonstrated that several exosome-specific signals colocalized with the early endosome markers EEA1 and Rab5, suggesting the internalization of exosomes via the endocytic pathway.23) In addition, flow cytometry analysis has shown a reduction in the BeWo cell-derived exosome uptake in MFSD2A- or ASCT2-silenced BeWo cells.23) These findings suggest that syncytin-1-ASCT2 and syncytin-2-MFSD2A interactions mediate the internalization of placenta-derived EVs into the target placental trophoblasts. This idea is supported by previous reports that EVs derived from a syncytin-1-transduced human prostate cancer cell line (PC3) are rapidly taken up by PC3 cells or human umbilical vein endothelial cells (HUVECs) in a syncytin-1- and ASCT2-dependent manner.33) This transport system of placenta-derived EVs could be applied for targeted drug delivery to treat placental syncytiotrophoblasts in diseases.
Placenta-derived EVs are released into the maternal circulation and subsequently interact with both maternal endothelial cells and maternal immune cells.
4.1. Endothelial CellsAn in vitro study demonstrated that a combination of cell surface binding, phagocytosis, and endocytosis is necessary for the rapid interactions between the human microvascular endothelial cell line (HMEC-I cells) and nanovesicles derived from a human first-trimester placenta.34) EVs derived from human placental syncytiotrophoblasts deliver placenta-associated microRNAs (miRNAs) to primary human coronary artery endothelial cells (HCAECs), resulting in the downregulation of specific target genes.35) This uptake of syncytiotrophoblast-derived EVs by HCAECs is mediated by clathrin, lipid rafts, and dynamin, and clathrin-mediated endocytosis is the major pathway involved36); however, the molecular mechanism underlying this uptake remains unclear.
4.2. Immune CellsThe human placenta is a fetus-derived organ that is always in contact with the maternal immune system during pregnancy. Therefore, the induction of immune tolerance is a crucial step for a successful pregnancy. In the past two decades, there has been an explosion of research on the role of placenta-derived EVs in modulating the function of maternal circulating and tissue-resident immune cells during healthy pregnancy. Several studies have investigated placenta-associated miRNAs and their functions.37–39) However, a lack of knowledge regarding the molecular transport of EVs to the immune cells remains. A previous report demonstrated that exosomes derived from Sw71 cells, a human first-trimester trophoblast cell line, were taken up by macrophages derived from human peripheral blood mononuclear cells isolated from normal non-pregnant donors. The uptake was completely blocked by pretreatment with cyochalasin D, an inhibitor of phagocytosis.40) However, little is known about the mechanism underlying internalization in other immune cells, such as lymphoid T cells and natural killer (NK) cells, although data from several in vitro and in vivo studies suggest that placenta-associated miRNAs are delivered into Jurkat cells (a human leukemic T-cell line) and maternal NK cells mediated by placenta-derived EVs. Kambe et al. demonstrated that human placenta-associated miR-517a-3p encapsulated in BeWo-derived exosomes was internalized by Jurkat cells and subsequently inhibited the mRNA expression of PRKG1 in Jurkat cells.41) Furthermore, using NK cells isolated from the peripheral blood in vivo, they found that circulating miR-517a-3p is delivered to maternal NK cells.41,42) Gene therapy is a therapeutic option used in modern medicine. Although immune cells are the most interesting targets for gene therapy, difficulties exist in clinical settings because of the low transfection efficiency of T lymphocytes and NK cells.43–45) The transport system of placenta-derived EVs can be used for gene transfection into hard-to-transfect immune cells.
During pregnancy, placenta-derived EVs are continuously released into the maternal circulation.6–8) Emerging data suggest that there are systems for transporting placenta-derived EVs into maternal organs, including the lungs and the liver.
5.1. Maternal LungAn in vivo study by Nguyen et al. demonstrated that EVs isolated from the plasma of pregnant mice and intravenously administered to non-pregnant mice were distributed in the lungs of the non-pregnant mice. In contrast, EVs isolated from the plasma of non-pregnant mice and administered to non-pregnant mice were not detected in the lungs of the non-pregnant mice. These results suggest that pregnancy-specific EVs could be distributed to the lungs, although plasma EVs obtained from non-pregnant mice could not be distributed to the lung.46) Nguyen et al. also demonstrated that EVs derived from the placenta of gestational day (GD)14.5 mice were detected in lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1)-positive and CD68-double-positive intestinal macrophages in the lung after being intravenously administered in non-pregnant mice.46) Immunoblot analysis demonstrated that EVs derived from GD14.5 mouse placenta expressed the integrins α3, αV, α5, β1, and β3. Furthermore, the signals of placenta-derived EVs in the mouse lungs were diminished when EVs were pre-incubated with the HYD-1 peptide, which can block the binding of integrin α3β1 and the pulmonary basement membrane glycoprotein laminin,47,48) prior to their intravenous administration in non-pregnant mice.46) These results suggest that the interaction between the integrin α3β1 in mouse placenta-derived EVs and laminin in the mouse pulmonary basement membrane is a potent mechanism for placenta-derived uptake of EVs by interstitial lung macrophages.
Proteomic analysis has shown that human placenta-derived EVs express several integrins.3,4,21) Tong et al. demonstrated that human placenta-derived microvesicles or nanovesicles administered to pregnant or non-pregnant mice were also trafficked to the lungs.34,49) Laminin is expressed in human lungs,50) raising the possibility that EVs derived from the human placenta could interact with the lungs in women during pregnancy.
5.2. Maternal LiverA mouse study showed that the placenta-derived EVs from GD14.5 pregnant mice could be transported to CD31-positive endothelial cells and F4/80-positive Kupffer cells in the liver of non-pregnant mice after being intravenously administered.46) Pre-incubation of the EVs with the RGD peptide, which blocks the integrins α5β1 and αVβ3 from binding to its receptor, fibronectin,51) inhibits trafficking to the liver.46) Another in vivo study demonstrated that human placenta-derived microvesicles or nanovesicles administered to pregnant or non-pregnant mice also localize to the liver34,49) and that fibronectin is enriched in the human liver and mouse liver.52,53) These findings suggest that the interaction of the integrins α5β1 and αVβ3 with placenta-derived EVs and fibronectin in the liver mediates EV transport. It has also been reported that plasma EVs isolated from pregnant mice and those isolated from non-pregnant mice were both detected in the livers of non-pregnant mice when these EVs were intravenously administered. These results suggest that the distribution of plasma EVs to the liver is not placenta- and/or pregnancy-specific manner.46)
A recent prospective neuroimaging study demonstrated that pregnancy leads to substantial changes in the structure of the human brain and primarily results in a reduction in the gray matter volume in regions associated with social cognition.54) Thus, the placenta has the potential to interact with the maternal brain; however, the underlying molecular mechanisms underlying this interaction remain unknown. In vivo mouse studies demonstrated that the localization of exogenous placenta-derived EVs to the brain was not detected in either pregnant34) or non-pregnant mice46) using representative whole-organ imaging analysis. In contrast, an in vitro transwell study demonstrated that the exposure of human-derived brain endothelial cell monolayers to EVs isolated from the plasma of women with preeclampsia increased permeability and reduced transendothelial electrical resistance.55) Furthermore, an in vivo mouse study demonstrated that the injection of EVs derived from human term placental explants subjected to hypoxia increased their permeability to Evans blue stain in the brains of healthy non-pregnant mice.55) These findings suggest that placenta-derived EVs have the potential to internalize in brain microvascular endothelial cells, which constitute the blood–brain barrier (BBB).
6.1. Blood–Brain BarrierThe BBB is a major limiting factor in the delivery of molecules to the central nervous system. Accumulating evidence, both in vivo (Table 1) and in vitro (Table 2), has shown that EVs can be internalized and cross the BBB. For instance, when exosomes derived from the human breast cancer MDA-MB-231 sub-cell line, which colonizes in the brain (831-BrT), were injected into healthy mice, immunofluorescence analysis using frozen brain tissue sections demonstrated that several exosome-specific signals colocalized with the brain endothelial cell marker CD31.19) Tominaga et al. demonstrated that EVs derived from brain-metastasized MDA-MB-231 cells (BMD2a and BMD2b) are taken up by Macaca virus brain endothelial cells and deliver miR-181c, thereby increasing the permeability of the BBB.56) They showed that the administration of BMD2b-derived EVs into the tail vein of mice promoted the metastasis of breast cancer cells to the brain and was preferentially incorporated into the brain in vivo.56) Furthermore, live imaging analysis using zebrafish embryos revealed that the movement of EVs derived from the brain-seeking MDA-MB-231 cells (Br-EVs) through the endothelial recycling endocytic pathway, can be monitored.57) However, little is known about the molecular transport mechanisms underlying cancer cell-derived EVs into the human BBB. To our knowledge, Kuroda et al. showed for the first time that exosomes derived from a human skin melanoma cell line (SK-MEL-28) could bind to the cell surface receptor CD46, which mediates the internalization of exosomes into human brain microvascular endothelial cells (hCMEC/D3), an in vitro model of the human BBB.58)
| Source of EVs | In vivo evaluation of EVs transport to the brain | References | ||||
|---|---|---|---|---|---|---|
| Origin (Cell types) | EVs collection | Labeling method | EVs administration (Species) | Evaluation | Distribution in the brain/Brain influx rate (Kin ± standard error) (µL/g·min) | |
| Brain-seeking MDA-MB-231 cell line (Human breast cancer) | U | TdTomato | Intracardiac injection (Zebrafish) | Live imaging of embryo | Crossing of the brain endothelium | 57) |
| MDA-MB-231 sub-cell line, which colonizes in the brain (831-BrT) (Human breast cancer) | U | PKH26 | Intracardiac injection (Mouse) | Imaging of brain frozen tissue sections | CD31-positive endothelial cells | 19) |
| Circulating blood (Mouse serum) | U | DiD | Intravenous injection (Mouse) | Imaging of brain frozen tissue sections | Striam, substantia nigla, and hippocampus | 62) |
| Z310 cells stably transfected with FRα (Rat choroidal epithelial cell) | U | PKH26/FA-FITC | Intraventricular cannula injection (Mouse) | Imaging of brain frozen tissue sections | GFAP-positive astrocytes, NeuN-positive neurons | 64) |
| Raw 264.7 (Mouse macrophage) | U | 125I | Injection to jugular vein (Mouse) | Integration plot | Kin: 0.32 | 60) |
| J774A.1 (Mouse macrophage) | MC | 125I | Injection to jugular vein (Mouse) | Integration plot | Kin: 0.083 ± 0.017 | 59) |
| NIH-3T3 (Mouse fibroblast) | Kin: 0.103 ± 0.040 | |||||
| Primary T cell (Human T cell) | Kin: 0.547 ± 0.116 | |||||
| SCCVII (Mouse oral squamous cell carcinoma) | Kin: 0.162 ± 0.015 | |||||
| MEL526 (Human melanoma) | Kin: 0.091 ± 0.035 | |||||
| MDA-MB-231 (Human breast cancer) | Kin: 0.098 ± 0.034 | |||||
| PCI-30 (Human head&neck cancer) | Kin: 0.057 ± 0.011 | |||||
| SCC-90 (Human head&neck cancer) | Kin: 0.123 ± 0.015 | |||||
| Kasumi (Human Leukemia) | Kin: 0.044 ± 0.013 | |||||
U, Ultracentrifugation; MC, Mini-size exclusion Chromatography.
| Source of EVs | Ligand on EVs | Target cells | Receptor on Target cells | Inhibitor used to identify receptor(s) | References | ||
|---|---|---|---|---|---|---|---|
| Origin (Cell types) | EVs collection | Labeling method | |||||
| SK-MEL-28 (Human melanoma) | MC | PKH67 | hCMEC/D3 | CD46 | CD46 siRNA | 58) | |
| SK-MEL-28 (Human melanoma) | EQ | PKH67 | hCMEC/D3 | RGD-receptor integrins | Antibodies against integrin α5 and αV, RGD peptide | ||
| Raw 264.7 (Mouse macrophage) | U | CM-DiI | Lymphocyte function-associated antigen 1 (LFA-1) | hCMEC/D3 | Intercellular adhesion molecule 1 (ICAM-1) | Antibodies against LFA-1 or ICAM-1 | 60) |
| C17.2 Neural stem cell (Mouse neural stem cell) | U | DiI | hCMEC/D3 | Heparan sulfate proteoglycans (HSPGs) | Heparin, Heparinase III | 61) | |
| Circulating blood (Mouse serum) | U | PKH67 | Transferrin | bEnd.3 | Transferrin receptor (TfR) | Transferrin | 62) |
MC, MagCapture (FUJIFILM Wako Pure Chemical Corporation); EQ, ExoQuick-TC (System Biosciences); U, ultracentrifugation.
In recent years, an increasing number of publications have reported that EVs derived from both cancerous and non-cancerous cells can cross the BBB. Banks et al. investigated the ability of EVs derived from mouse/human cancerous and noncancerous cell lines to cross the mouse BBB using capillary depletion method.59) They found that all EVs could cross the mouse BBB and that they crossed at varying rates; however, the molecular mechanism of EV transport remains unknown.59) In another in vivo study, performed by Yuan et al., 94% of 125I-exosomes derived from a mouse macrophage cell line (RAW 264.7) penetrated the mouse brain and accumulated in the parenchymal fraction when healthy mice were injected with 125I-exosomes via the jugular vein.60) They also observed a significant net influx of 125I-macrophage-derived exosomes in the brain: the slope of the delta brain/serum ratio was estimated to be 0.32 µL/g·min.60) Furthermore, in vitro studies have demonstrated that exosomes derived from RAW 264.7 cells can be internalized into the human brain microvascular endothelial cell line (hCMEC/D3) and that the uptake was significantly decreased in the presence of the antibodies of lymphocyte function-associated antigen 1 (LFA-1) or intercellular adhesion molecule 1 (ICAM-1), indicating that the interaction of LFA-1 on exosomes and ICAM-1 on the cell surface mediates the uptake of exosomes by hCMEC/D3 cells.60)
Another in vitro uptake study demonstrated that hCMEC/D3 cells transport exosomes derived from mouse C17.2 natural stem cells via endocytosis, wherein heparan sulfate proteoglycans (HSPGs) act as receptors.61) Moreover, when exosomes isolated from the serum of healthy mice were injected into healthy mice, immunofluorescence analysis using frozen brain tissue sections revealed exosome-specific signals in the striatum, substantia, and hippocampus.62) When the mouse brain microvascular endothelial cell line (bEnd.3) was pre-incubated with transferrin for 30 min and then mixed with exosomes from the mouse serum, the uptake of exosomes was significantly reduced, suggesting that the transferrin-transferrin receptor (TfR) interaction is involved in the uptake mechanism.62) The transferrin-TfR interaction at the BBB is supported by a previous study, which demonstrated that liposomes conjugated with a high-affinity antibody for the TfR were taken up by the in vitro human BBB and in vivo rat BBB via TfR-mediated endocytosis.63) Taken together, the rodent BBB and human BBB express several receptors involved in EV transport, such as CD46, ICAM-1, HSPGs, and TfR. Human placenta-derived EVs express serotransferrin and lactotransferrin,3,20) suggesting that placenta-derived EVs have the potential to cross the BBB via TfR-mediated endocytosis.
6.2. Blood–Cerebrospinal Fluid Barrier (BCSFB)Emerging in vivo evidence has shown that EVs can bypass the BBB and cross the blood-cerebrospinal fluid barrier (BCSFB). The BCSFB also plays an integral role in the brain barrier function, together with the BBB. The BBB is established by brain microvascular endothelial cells, whereas the BCSFB is established by the choroid plexus epithelial cells. When folate receptor-α (FRα)-containing exosomes were administered to healthy mice via intraventricular cannulation, immunofluorescence analysis using frozen brain tissue sections revealed that several exosome-specific signals were detected in glial fibrillary acidic protein (GFAP)-positive astrocytes and NeuN-positive neurons.64) In contrast, FRα-negative exosomes were unable to cross the ependymal cell layer.64)
This review summarizes the molecular mechanisms underlying placenta-derived EV transport. Accumulating data on EV transport to the placenta and maternal organs will certainly be helpful in the design of optimal drug candidates. For instance, EV-based delivery has the potential to be used for drug delivery of peptides and nucleic acids. The development of an inhibition system for the delivery of EVs derived from abnormal placentas may have potential as therapeutic agents for pregnancy-related complications. However, numerous challenges remain in the targeted transport of placenta-derived EVs. As EVs are highly heterogeneous, more than one route is considered to be involved in EV transport. Furthermore, most recent findings on EV transport have been obtained in rodents, and extrapolation of experimental data from rodents to humans remains unclear. In recent years, an increasing number of publications have reported proteomic profiling of various human tissues, as well as placenta-derived EVs. Such data would enable us to reconstruct the in vivo human transport activity of EVs. Research on EVs is continuously expanding and developing. Therefore, we believe that EV-based delivery of large molecule drugs based on a further understanding of the molecular transport mechanisms of EVs in humans will be achieved in the near future.
This study was supported in part by JSPS KAKENHI (Grant numbers 20K22708 and 21K15314). This work was also supported in part by the Home for Innovative Researchers and Academic Knowledge Users (HIRAKU) program and a research program for the development of an intelligent Tokushima artificial exosome (iTEX) in Tokushima University.
The authors declare no conflict of interest.
