Feasibility of placenta-derived mesenchymal stem cells as a tool for studying pregnancy-related disorders

The cellular and molecular mechanisms responsible for pregnancy-related disorders remain unclear. We investigated the feasibility of using placenta-derived mesenchymal stem cells (MSCs) as a tool to study such pregnancy-related disorders. We isolated and expanded adequate numbers of cells with characteristic features of MSCs from the chorionic plate (CP-MSCs), chorionic villi (CV-MSCs), and decidua basalis (DB-MSCs) of human term placental tissues. All placenta-derived MSCs expressed pregnancy-associated C14MC microRNA (miRNA) (miR-323-3p). Interestingly, the placenta-specific C19MC miRNAs (miR-518b and miR517a) were clearly expressed in CP-MSCs and CV-MSCs of foetal origin, but were barely expressed in DB-MSCs of maternal origin. Furthermore, expression levels of placenta-specific C19MC miRNAs in CV-MSCs remained stable during the ex vivo expansion process and across different pregnancy phases (first trimester versus third trimester). High-efficiency siRNA transfection was confirmed in twice-passaged CV-MSCs with little toxicity, and microarray analysis was used to screen for miR-518b target genes. Placenta-derived MSCs, especially CV-MSCs, are a potential tool for investigating the role of placental miRNAs in pregnancy-related disorders.

Scientific RepoRts | 7:46220 | DOI: 10.1038/srep46220 placental development and maintenance. These results indicate the potential use of placenta-derived MSCs as a tool for investigating the mechanisms behind pregnancy-related disorders.
In this study, we aimed to expand MSCs from different parts of human term placentas including the chorionic plate (CP) and chorionic villi (CV) of foetal origin, and the decidua basalis (DB) of maternal origin (Fig. 1), and to examine the expression of pregnancy-associated miRNAs in MSCs primarily expanded from placental tissues. We also investigated the transfection efficiency of CP-MSCs and CV-MSCs from term placentas with small interfering RNAs (siRNAs), and miR-518b target genes were further screened for by microarray analysis. These preliminary results suggest that CV-MSCs are a potential tool for investigating the role of placental miRNAs in pregnancy-related disorders.

Results
Propagation and characterisation of placenta-derived MSCs. We separated different parts (CP, CV, and DB) of human third-trimester placental tissues and cultured explants to generate MSCs (Fig. 1a). All cells that grew out from CP, CV, and DB tissues were morphologically fibroblast-like cells (Fig. 1b). The average population-doubling times of CP-MSCs, CV-MSCs, and DB-MSCs at passage two were 42.8 ± 6.3, 37.0 ± 1.4, and 68.1 ± 30.3 h, respectively (Fig. 1c). The origin of the MSCs was examined by microsatellite genotyping of the short tandem repeat (STR) markers D8S1179, TH01, vWs, and amelogenin (Fig. 1d), which revealed the biparental origins of CP-MSCs and CV-MSCs, compared with maternal-origin DB-MSCs.
The phenotypes of the MSCs were examined by flow cytometry and immunocytochemistry (Fig. 2). Flow cytometry analysis showed that all CP-MSCs, CV-MSCs, and DB-MSCs expressed the MSC markers CD44, CD73, CD90, and CD105 but not the haematopoietic cell markers CD34 and CD45 (Fig. 2a). CP-MSCs, CV-MSCs, and DB-MSCs also expressed human leukocyte antigens (HLA)-A, B, and C (MHC class 1 cell surface receptors) but not HLA-DR (MHC class 2 cell surface receptor) (Fig. 2a), or HLA-G (MHC class 1 cell surface receptor, which is known to be expressed in extravillous trophoblasts) (Fig. 2a). According to immunocytochemistry analysis, all the MSCs primarily expanded from different parts of the placenta tissue were negatively stained for the pan-trophoblast-specific marker keratin 7 but positively stained for the mesenchymal marker vimentin (Fig. 2b).
We also compared the expression levels of pregnancy-associated miRNAs between primarily expanded MSCs and their original placental tissues. The relative expression levels of miR-323-3p varied between primarily expanded MSCs and their original tissues (Fig. 4a), while expression levels of C19MC miRNAs (miR-518b and -517a) were significantly higher in CP and CV tissues than in twice-passaged CP-MSCs and CV-MSCs (p < 0.001 vs. CP tissues or CV tissues, respectively) (Fig. 4b,c).
Expression of pregnancy-associated miRNAs in placental MSCs remained stable through different pregnancy phases and during ex vivo expansion. As a potential tool for investigating the mechanisms responsible for pregnancy-related disorders, it is essential to understand the stability of pregnancy-associated miRNA expression levels in these placenta-derived MSCs. We therefore determined if the expression of pregnancy-associated miRNAs changed during the ex vivo expansion process. The expression of miR-323-3p in CP-MSCs and CV-MSCs was significantly higher in later (p8) compared with earlier passage (p2) cells (p < 0.001 vs. P8) (Fig. 5a), while levels of miR-518b and miR-517a remained relatively stable during ex vivo expansion (Fig. 5b,c).
We also primarily expanded and compared the properties of first-and third-trimester CV-MSCs, and showed that they had similar morphological features and cellular properties (Fig. 6a,b). However, first-trimester CV-MSCs showed relatively higher expression of miR-323-3p compared with third-trimester CV-MSCs, though the difference was not significant (Fig. 6c), while miR-518b and -517a levels were similar in cells from both trimesters (Fig. 6c). We determined the effects of siRNA transfection in twice-passaged CP-MSCs and CV-MSCs derived from third-trimester placenta. siRNA transfection had little toxic effect (Fig. 7a). Furthermore, expression levels of miR-518b and miR-323-3p were effectively facilitated and suppressed by transfection with miRNA mimic and miRNA inhibitor, respectively (Fig. 7b).

Discussion
The cytological characteristics of MSCs vary according to their origin 40 . Placenta-derived MSCs were recently reported to possess better immunoregulatory properties than umbilical cord-derived MSCs 41 . Furthermore, MSCs from the amnion, chorion, and umbilical cord of human placenta tissues have different gene expression profiles and differentiation capacities 42 , suggesting the existence of heterogeneity among MSCs originating from different tissues. However, the biological features of MSCs in placental tissues and their role in regulating placental development remain poorly understood. The present study aimed to investigate the feasibility of using placenta-derived MSCs as a tool to study the mechanisms responsible for pregnancy-related disorders.
We initially expanded MSCs from placenta tissues by seeding tissue explants from different parts of the placenta onto culture dishes. These explants produced adequate numbers of cells with high proliferative potency. Numerous methods have been used to isolate/expand MSCs from placental tissue, including enzymatic digestion of tissues to harvest MSCs as a single-cell suspension for further cell expansion 43,44 , or by seeding tissue fragments, as in the current study. Contamination between maternal-and foetal-origin MSCs remains a problem with expansion of MSCs from placental tissues 45,46 , but we were able to generate pure cells without contamination. The explant culture method also has the advantages of maintaining stemness and retaining identical cell properties over time 47,48 .
Considering the distinct role of pregnancy-associated miRNAs in placental development, we investigated the expression levels of the major clusters of pregnancy-associated miRNAs, C19MC and C14MC, in these MSCs primarily expanded from human placental tissues. C19MC miRNAs are placenta-specific miRNAs regulated by genomic imprinting, with only the paternally inherited allele being expressed in the placenta 49,50 . In contrast, C14MC miRNAs are expressed in both embryonic and placental tissues 51,52 , and are generally accepted as pregnancy-associated, rather than placenta-specific miRNAs 2,3 . C14MC miRNAs (miR-323-3p) were expressed in all the placenta-derived MSCs, including CP-MSCs, CV-MSCs, and DB-MSCs, and also in WJ-MSCs and UCB-MSCs. In contrast, the placenta-specific C19MC miRNAs (miR-518b and miR517a) were only clearly detected in CP-MSCs and CV-MSCs, with low or absent expression in DB-MSCs, WJ-MSCs, and UCB-MSCs.
In accord with a previous study 2 , expression levels of C14MC miRNAs were lower in third-compared with first-trimester CV-MSCs, and decreased with pregnancy progression. Interestingly, expression levels of the placenta-specific C19MC miRNAs were comparable in first-and third-trimester CV-MSCs, and remained stable during the ex vivo expansion process (within approximately 60 days). We also compared the expression of C19MC miRNAs between primarily expanded MSCs and their original tissues, and showed lower expression levels in CP-MSCs and CV-MSCs compared with the equivalent original tissues. Given that C19MC miRNAs are highly expressed in trophoblast cells 53 , it is not surprising that they were less enriched in placenta-derived MSCs compared with their original tissues.
Placenta-derived MSCs are expected to be useful tools for studying pregnancy-related disorders because they are easily expanded ex vivo, they express specific pregnancy-associated miRNAs, and they show high transfection efficiencies for siRNAs. In this study, we only expanded MSCs from placental tissues from uncomplicated pregnancies. However, we also revealed different characteristics among CP-MSCs, CV-MSCs and DB-MSCs, supporting the idea of heterogeneity and tissue-specificity among placental MSCs for cell properties and proliferative capacities 13,42,44,54 . We have started to expand and characterize placental MSCs from abnormal conditions (e.g. preeclampsia and/or foetal growth restriction); however, there is much to learn to understand the functional roles and detailed mechanisms of action of pregnancy-associated miRNAs in pregnancy-related disorders.
In summary, we successfully expanded CP-MSCs, CV-MSCs, and DB-MSCs from placental tissue, resulting in cells with high proliferative potency and clear expression of pregnancy-associated miRNAs. The expression of placenta-specific C19MC miRNAs in CV-MSCs remained stable throughout different pregnancy phases and during ex vivo expansion. Placenta-derived MSCs, especially CV-MSCs, represent a potentially useful tool, not only for mechanistic understanding, but also for the treatment of pregnancy-related disorders 16,[55][56][57] .

Ethics. This study was approved by the Institutional Review Boards for Ethical, Legal and Social Issues in
Nagasaki University Graduate School of Biomedical Sciences (13052715). All samples were obtained after receiving written informed consent. The experiments were performed in accordance with the institutional and national guidelines.
Primary isolation and expansion of human placental MSCs. We collected third-trimester (n = 3) placentas after elective caesarean section at 38-39 weeks of gestation, and first-trimester placentas (n = 3) after elective pregnancy termination at 8-10 weeks of gestation. We performed ex vivo expansion using explant methods, as described previously, with minor modifications 58,59 . Briefly, placental tissues were collected immediately after delivery and stored in Hank's balanced salt solution (Life Technologies, Carlsbad, CA, USA) at 4 °C. After extensive washing in Dulbecco's phosphate-buffered saline and mechanical removal of the amnion and blood vessels, third-trimester placental tissues were separated into CP, CV, and DB, but only CV was isolated from first-trimester placentas, because of the difficulty in discriminating between CP and DB. The tissues were cut into small pieces (1-2 mm) and cultured as explants on 6-cm culture dishes coated with 10 μ g/ml human fibronectin (Corning, Corning, NY, USA). Within 1 week, fibroblast-like cells grew out from the tissue fragments, and became confluent at approximately 2 weeks. These cells were then collected using 0.25% trypsin-EDTA (Gibco, Waltham, MA, USA) and passaged for cell expansion. All cultures were incubated in a 5% CO 2 incubator at 37 °C in Dulbecco's modified Eagle's medium (DMEM) (Wako, Osaka, Japan) supplemented with 10% foetal bovine serum (Hyclone Laboratories, Logan, UT, USA), 10 ng/ml human recombinant basic fibroblast growth factor (Wako), and 1% penicillin (100 U/ml)/streptomycin (100 U/ml) solution (Life Technologies).
Culture of other cells. The BeWo trophoblast cell line, derived from a human gestational choriocarcinoma, was obtained from the Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan) and maintained in Ham's F-12 medium (Wako) supplemented with 15% foetal bovine serum and 1% penicillin/streptomycin. UCB-MSCs and WJ-MSCs used as other-placental site-derived MSC models originated from the foetus, but not placenta, were kindly gifted by Dr. Doi 32 and maintained in DMEM supplemented with 10% foetal bovine serum and 1% penicillin/streptomycin. Cells were cultured in a 5% CO 2 incubator at 37 °C.
Cell growth assay. Twice-passaged CP-MSCs, CV-MSCs, and DB-MSCs were seeded onto 6-well plates at a density of 5.2 × 10 3 cells/cm 2 . The cells were then collected as single-cell suspensions at 3, 5, and 7 days, respectively. The total numbers of collected cells were counted with a haemocytometer to evaluate cell proliferation.
Genotyping of STR polymorphisms. Genomic DNA was extracted using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) from cultured MSCs, umbilical cord (baby) and parent peripheral blood samples. STR loci were analysed using the Powerplex 16 system (Promega, Madison, WI, USA) using an ABI PRISM3500 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) and GeneMapper software (Applied Biosystems), following the manufacturer's instructions. qRT-PCR. Total RNAs containing small RNA molecules were collected using mirVana miRNA Isolation Kit (Ambion, Waltham, MA, USA), according to the manufacturer's instructions. Total RNAs were extracted from MSCs and from the original placental tissues for comparison. Total RNA concentrations were measured using NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA). A total of 100 ng RNA was used for the following the steps. Five specific primers and TaqMan probes for test and control miRNAs (C19MC miRNAs miR-518b (assay ID 001156) and miR-517a (assay ID 002402); C14MC miRNA 323-3p (assay ID 002227); miR-21 (assay ID 000397); and U6 snRNA (assay ID 001973)) were used for TaqMan MicroRNA Assays (Applied Biosystems). Absolute qRT-PCR of miRNAs was performed as described previously [7][8][9] . For each miRNA assay, a calibration curve was prepared by 10-fold serial dilution of single-stranded cDNA oligonucleotides corresponding to each miRNA sequence from 1.0 × 10 2 to 1.0 × 10 8 copies/ml. Each sample and each calibration dilution were analysed in triplicate. The lower limit of detection for each assay was 300 RNA copies/ml [7][8][9] . Each batch of amplifications included three water blanks as negative controls for each of the reverse transcription and PCR steps. All the data were collected and analysed using a LightCycler ® 480 real-time PCR system (Roche, Basel, Switzerland).
Expression levels were represented as relative ratios using U6 snRNA as an endogenous control for normalisation.
Transfection of siRNA. Cells were transfected with synthetic miRNA mimic and miRNA inhibitor of hsa-miR-518b (assay ID MC12660 and MH12660) and -323-3p (assay ID MC12418 and MH12418) (Applied Biosystems), or with scramble controls (Ambion), using Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. A total of 30 nM of siRNA duplex was used for each transfection. Transfection efficiency was assessed by qRT-PCR.
Microarray analysis. Total RNAs were collected from the control MSCs (untreated twice-passaged CV-MSCs) and twice-passaged CV-MSCs 72 h after transfection with an miRNA mimic of miR-518b. The integrity of total RNAs was estimated using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Total RNAs (50 ng) of each MSC were labelled using a Low Input Quick Amp Labeling Kit (Agilent Technologies). miR-518b target genes were identified using a SuperPrint G3 human Microarray 8 × 60 ver. 3 (Agilent Technologies). The resulting intensity was normalized to the 75 percentile shift and the processed data was filtered for over 2-fold up-or down-regulation compared with control MSCs by GeneSpring Gx software ver. 13 (Agilent Technologies). Registered genes in the HUGO (Human Genome Organization) Gene Nomenclature Committee database (http:// www.genenames.org/) are listed as gene symbols in Supplemental Tables 1 and 2.
Statistical analyses. Data were shown as the mean ± standard error. Statistical significance was determined using Mann-Whitney U tests (SPSS ver. 23, IBM, Armonk, NY, USA). A p value < 0.05 was accepted as statistically significant.