Animals
C57BL/6 mice and CB17/Icr-Prkdc scid/CrlCrlj (SCID) mice were used in this study. All animals were treated according to the regulations of the Standards for Human Care and Use of Laboratory Animals of Tohoku University. The animal experiments were approved by the Animal Care and Experimentation Committee of Tohoku University Graduate School of Medicine (permission No. 2019MdA-265-03).
Preparation of human Muse cells
The Ethics Review Board approved the use of h-UC according to the guidelines of the Ethics Committee of the Graduate School of Medicine, Tohoku University (approval number: 2020-1-293). The h-UCs were obtained from 5 infants delivered at 35 to 38 weeks of gestation with parental written consent at the Department of Pediatrics, Kobe University Graduate School of Medicine (approval number: 1370).
Human-UC-MSCs were first isolated from the entire h-UCs using the explant culture method [26]. Cells were maintained in culture medium comprising α-minimum essential medium (α-MEM, MilliporeSigma, St Louis, Mo, USA) with 10% (vol/vol) fetal bovine serum (FBS, HyClone, Logan, UT, USA), 1 ng/mL basic fibroblast growth factor (Miltenyi Biotec, Bergisch Gladbach, Germany), 2 mM GlutaMAX I (ThermoFisher Scientific, Waltham, MA, USA), and 0.1 mg/mL kanamycin sulfate (ThermoFisher Scientific) at 37°C in 95% air and 5% CO2. Human-BM-MSCs, human adipose tissue-derived stem cells (h-ADSCs), and normal human dermal fibroblasts (NHDFs) were purchased from Lonza (Basel, Switzerland; h-BM-MSCs: PT-2501, h-ADSC: PT-5006, NHDF: CC-2511). Cells from passages 7 through 10 were used for Muse cell isolation.
Muse cells were isolated by fluorescence-activated cell sorting (FACS; BD FACS Aria™ II cell sorter, BD Biosciences [BD], San Jose, CA, USA). To isolate h-UC-SSEA-3 (+) cells, h-UC-MSCs were incubated with anti-SSEA-3 antibody (1:200 catalog #MA1-020; ThermoFisher Scientific) and stained with secondary antibody to fluorescein isothiocyanate (FITC)-conjugated anti-rat IgM (1:100, #112-095-075; Jackson ImmunoResearch Laboratories, West Grove, PA, USA).
To collect Muse cells, h-BM-MSCs, h-ADSCs, and NHDFs were incubated with anti-SSEA-3 antibody (1:1000, #330302; BioLegend, San Diego, CA, USA) and stained with secondary antibody FITC-conjugated anti-rat IgM (1:100) as previously described [27]. Human-UC-, h-BM-, h-AT-, and h-dermal tissue (h-DT)-Muse cells were then collected as SSEA-3(+) fractions by FACS [1].
Muse cells and non-Muse cells stably expressing either green fluorescent protein (GFP), mCherry, or Akaluciferase (Akaluc) were prepared as described previously [28]. Akaluc/pcDNA3, provided by Dr. Satoshi Iwano [29], was inserted into a vector to construct the pWPXL vector. In brief, either pWPXL-GFP for lentivirus-GFP, pWPXL-mCherry for lentivirus-mCherry, or pWPXL-Akaluc for lentivirus-Akaluc were mixed with pMD2G and pCMV deltaR8.74 and transfected into LentiX-293 T packaging cells (Takara Bio, Shiga, Japan) using Lipofectamine 2000 (ThermoFisher Scientific). After 3 days, the viral supernatant was collected, centrifuged, and filtered through a 0.45-µm filter. To collect GFP- and Akaluc-labeled h-UC-SSEA-3(+) cells, h-UC-MSCs were infected with lentivirus, and then SSEA-3(+) cells were separated by FACS using allophycocyanin (APC)-conjugated goat anti-rat IgM (1:100) as the secondary antibody.
Immunohistochemistry
To detect SSEA-3(+) cells in the h-UC, the tissue was fixed with 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB) overnight, embedded in paraffin, and cut into 3-µm thick sections. After deparaffinization and rehydration, sections were washed in phosphate-buffered saline (PBS), treated with 0.3% hydrogen peroxide (Fujifilm) in methanol (Fujifilm) for 30 min to inactivate the intrinsic peroxidase activity, and then incubated with 20% Block-Ace (KAC Co.; Kyoto, Japan)/5% bovine serum albumin (BSA, Nacalai Tesque, Kyoto, Japan)/0.3% Triton X-100 Fujifilm)/PBS solution for blocking. The sections were incubated overnight at 4℃ with anti-SSEA-3 antibody (1:200, #MA1-020; ThermoFisher Scientific) diluted with 5% Block-Ace/1% BSA/0.3% Triton X-100/PBS solution. After washing with PBS, the sections were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rat IgM antibody (1:200). Expression of SSEA-3 was visualized by 3,3'-diaminobenzidine tetrahydrochloride (Fujifilm) and counterstained with Mayer’s hematoxylin (Fujifilm).
To evaluate the differentiation of h-UC-SSEA-3(+) cells and h-BM-Muse cells in mouse injured liver, mice anesthetized with isoflurane (Fujifilm) were transcardially perfused and fixed with 4% PFA in 0.1 M PB. The liver was dissected out, postfixed with 4% PFA in 0.1 M PB for 24 h, embedded in Optimal Cutting Temperature (OCT) compound (Sakura FineTechnical Co., Ltd., Tokyo, Japan), and then cut into 6-µm-thick frozen sections. STEM121 was used as a human cytoplasmic-specific marker to detect human-derived cells in the mouse liver. Albumin was used as a hepatocyte marker. Lymphatic vessel endothelial hyaluronan receptor 1 (Lyve-1) was as a liver sinusoidal endothelial cell marker. After washing with PBS, frozen sections were incubated with 20% Block-Ace/5% BSA/0.3% Triton X-100/PBS solution, and then incubated with mouse monoclonal antibody STEM121 (1:100, #Y40410; TaKaRa Bio Inc) followed by secondary antibody with Alexa Fluor 488-conjugated donkey anti-mouse IgG antibody (1:200, #715-546-150; Jackson ImmunoResearch). After washing with PBS, frozen sections were first incubated with goat anti-albumin (1:100, #A80-229A; Bethyl Laboratories, Montgomery, TX, USA) or rabbit anti-Lyve-1 antibodies (1:100, #NB600-1008; Novus Biologicals, Centennial, CO, USA), then with the secondary antibody. The primary and secondary antibodies used in this experiment are listed in Supplemental Table 1. The sections were counterstained with 4',6-diamidino-2-phenylindole (DAPI; 1:500, D1306; ThermoFisher Scientific). Images were acquired by confocal laser microscopy (A1; Nikon). Ten randomly obtained images in each section were processed by ImageJ.1.53t to obtain a positive ratio for the albumin and Lyve-1 in STEM121(+) cells.
Quantitative reverse transcriptase- polymerase chain reaction (qPCR)
Total RNA was collected using the NucleoSpin® RNA XS (Macherey-Nagel, Duren, Germany), and cDNA was synthesized using Oligo(dT) 20 primer (ThermoFisher Scientific) and SuperScript® III reverse transcriptase (ThermoFisher Scientific). DNA was amplified with the Applied Biosystems 7500 Fast real-time PCR system according to the manufacturer’s instructions. Data were processed using the ΔΔCT method[30]. Pluripotent markers were evaluated using a SYBR Green expression assay with the following primers: Nanog-F: 5’-CAGCTCGCAGACCTACATGA-3’, Nanog-R: 5’-CTCGGACTTGACCACCGAAC-3’; POU5F1-F: 5’-AACCCACACTGCAGCAGATCA-3’, POU5F1-R: 5’-ACACTCGGACCACATCCTTC-3’; SOX2-F: 5’-TCCAACATCCTGAACCTCAGC-3’, SOX2-R: 5’-TCTGCGTCACACCATTGCT-3’; ACTB-F: 5’-CATGTACGTTGCTATCCAGGC-3’, and ACTB-R: 5’-CTCCTTAATGTCACGCACGAT-3’. Trophectoderm markers were evaluated using TaqMan gene expression assay with the following primers: caudal type homeobox 2 (CDX2; Hs01078080_m1), inhibitor of DNA binding 2 (ID2; Hs04187239_m1), tumor protein p63 (TP63; Hs00978343_m1), glial cells missing-1 (GCM1; Hs00961601_m1), endogenous retrovirus group FRD member 1 (ERVFRD-1; Hs01652148_m1), and placental growth factor (PLGF; Hs00182176_m1). Germline-lineage markers were evaluated with the following primers: B lymphocyte-induced maturation protein 1 (BLIMP1; Hs00153357_m1), nanos C2HC-type zinc finger 3 (NANOS3; Hs00928455_s1), developmental pluripotency associated 3 (DPPA3; Hs01931905_g1), and SSEA-1 (Hs01106466_s1). Hematopoietic lineage markers were evaluated with the following primers: fetal liver kinase 1 (FLK-1; Hs00911700_m1), TIE2 (Hs00945149_m1), CD31 (Hs00169777_m1), c-KIT (Hs00174029_m1), CD34 (Hs02576480_m1), and protein tyrosine phosphatase receptor type C (PTPRC; Hs04189704_m1). ACTB (Hs99999903_m1) was used as an endogenous internal control.
Surface marker expression
The antibodies used for double-staining with SSEA-3 were APC-labeled mouse anti-human HLA-ABC (1:100, #311409; BioLegend) and HLA-DR (1:100, #307609; BioLegend), phycoerythrin (PE)-labeled mouse anti-human CD29 (1:100, #556049; BD), CD90 (1:100, #555596; BD), CD44 (1:100, #550989; BD), CD73 (1:100, #550257; BD), CD166 (1:100, #559263; BD), CD34 (1:100, #555822; BD), CD45 (1:100, #555483; BD), and CD271 (1:100, #557196; BD). Purified mouse anti-human CD105 (1:100, #555690; BD) and von Willebrand factor (vWF, 1:100, #555849; BD) antibodies were detected with Pacific Blue-labeled goat anti-mouse IgG (1:100, #P31582; ThermoFisher Scientific). Purified mouse anti-human CD133 (1:100, #130-090-422; Miltenyi Biotec), SSEA-4 (1:100, #330401; BioLegend), and HLA-G (1:50, #557577; BD) primary antibodies were detected by APC-labeled goat anti-mouse IgG (1:100, #115-136-146; Jackson ImmunoResearch Laboratories). These cells were analyzed by FACS Aria™ II (BD).
Immunocytochemistry
Immunocytochemistry was performed as previously described [1]. Single-cell suspension-derived clusters generated from h-UC-SSEA-3(+) cells were fixed with 4% PFA in 0.1 M PB, embedded in OCT compound, and then cut into 6-µm-thick frozen sections. Differentiated cells derived from the h-UC-SSEA-3(+) cell cluster were grown in gelatin-coated dishes. Cells were fixed with 4% PFA in 0.1 M PB. After washing with PBS, the samples were incubated with 20% Block-Ace/5% BSA/0.3% Triton X-100/PBS solution for blocking, primary antibodies diluted in 5% Block-Ace/1% BSA/0.3% Triton X-100/PBS solution overnight at 4℃, washed 3 times with PBS, and then incubated with 0.2% Triton X-100/PBS solution containing secondary immunofluorescent antibodies (1:200) for 1 h at room temperature. After washing with PBS, the samples were stained with DAPI (1:500). The primary and secondary antibodies used in this experiment are listed in Supplemental Tables 1 and 2. Images were acquired by a confocal laser microscope (A1; Nikon).
Droplet digital telomere repeat amplification protocol
The telomerase extension reaction was measured by droplet digital telomere repeat amplification protocol (ddTRAP) [31]. Whole-cell lysates were prepared in NP-40 lysis buffer (10 mM Tris-HCl [pH 8.0], 1 mM MgCl2, 1 mM EDTA, 1% [vol/vol] NP-40, 25 mM sodium deoxycholate, 10% [vol/vol] glycerol, 150 mM NaCl, 5 mM β-mercaptoethanol, and 0.1 mM AEBSF [4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride]), and added to extension reaction containing 1× TRAP reaction buffer (10× concentration: 200 mM Tris-HCl [pH 8.3], 15 mM MgCl2), 0.4 mg/mL BSA, 200 nM telomerase substrate primer (TS primer; 5′-AATCCGTCGAGCAGAGTT), and deoxynucleotide triphosphates (2.5 mM each), and incubated for 40 min at 25°C followed by telomerase inactivation at 95°C for 5 min, and then held at 4°C. The ddPCR reaction contained 1× EvaGreen ddPCR Supermix (Bio-Rad, Hercules, CA, USA), 50 nM TS primer, 50 nM ACX primer, and an extension product. Droplets were produced in the QX200 droplet generator (Bio-Rad). PCR was performed on a C1000 thermocycler (Bio-Rad) with a ramp rate of 2.5°C/s between all steps. Activation of Taq polymerase (95°C for 5 min) was followed by 40 cycles of 95°C for 30 s, 54°C for 30 s, and 72°C for 30 s, then held at 12°C. Following PCR, fluorescence was read on the QX200 droplet reader (Bio-Rad). Data were analyzed using QuantaSoft software. The threshold between positive and negative droplets was determined by including controls such as a no-template lysis buffer control. HeLa cells were used as a positive control.
Preparation of mouse embryonic stem cells
Mouse ESCs (TT2 cells) were cultured at 37℃ in Dulbecco's modified Eagle's medium (DMEM, ThermoFisher Scientific) containing 15% FBS, 0.1 mg/mL kanamycin sulfate, 0.1 mM MEM non-essential amino acid solution (ThermoFisher Scientific), 1 mM sodium pyruvate solution (ThermoFisher Scientific), 1000 U/mL leukemia inhibitory factor (LIF, MilliporeSigma), and 100 mM 2-mercaptoethanol (Nacalai Tesque) on mitomycin-C (Nacalai Tesque)-treated mouse embryonic fibroblast (MEF) feeder cells established from 12.5-day embryos of C57BL/6 mice, as previously described [3].
Test for teratoma formation in immune-deficient mice testes
Human-UC-SSEA-3(+) cells (1×105 cells) were suspended in PBS and injected into the testes of 8-week-old SCID mice (n = 6; Charles River Laboratories Japan, Yokohama, Japan) using a glass micropipette under anesthesia with isoflurane. Mice were killed by a lethal dose of isoflurane anesthesia at 6 months after injection. For controls, PBS (n = 2) was injected into the testes as a negative control and 5×105 mouse ESCs (n = 4) were injected as a positive control; all were killed at 12 weeks after injection. Tissues were fixed with 4% PFA in 0.1 M PB, and 3-µm-thick paraffin sections were analyzed by hematoxylin-eosin staining.
Migration assay
Human-UC-SSEA-3(+) cells and h-BM-Muse cells were prepared by cell sorting as described above. A Matrigel invasion chamber (BD) was used and the experiment was performed according to the manufacturer’s protocol. Serum-free medium alone or liver tissues from normal C57BL/6 mice in α-MEM were placed in the lower chamber. Human-UC-SSEA-3 (+) or h-BM-Muse cells (2.5×104) were placed into the upper chamber and incubated at 37°C in 5% CO2 for 24 h. Migrated cells were fixed with 4% PFA in 0.1 M PB for 15 min and stained with Mayer’s hematoxylin (Fujifilm). The number of migrated cells in each sample was counted under a 20x objective in 4 fields, and the mean of 3 samples was calculated using ImageJ.1.53t.
Preparation of dead cell fragments
Mouse-hepatic cell line Hepa1-6 (m-Hepa1-6) cells were infected with mCherry lentivirus and then expanded. To generate apoptotic-m-Hepa1-6, 2×106 cells were treated with 100 µM rotenone (MilliporeSigma) and 100 µM antimycin A (MilliporeSigma) in DMEM for 24 h [9]. More than 95% of cells died and floated in the culture medium. The dead cell fragments were vortexed, filtered with a 40-µm cell strainer (SPL, Life Sciences, Pocheon, Korea), collected by centrifugation at 800×g for 5 min, and stored in CELLBANKER 1 Plus (Juji-Field, Tokyo, Japan) at -80℃ until use.
In vitro live image observations of phagocytosis
Incubation of GFP-h-UC-SSEA-3(+) cells and apopototic-mCherry-m-Hepa1-6 dead cell fragments was recorded under a laser confocal microscope (Nikon) at 37°C in 5% CO2 for live image analysis. The cells were counterstained with Hoechst 33342 (ThermoFisher Scientific). Images were captured every 15 min with a Z-series for three-dimensional imaging using a confocal laser microscope (A1; Nikon). Bitplane Imaris 9.7.2 software (Oxford Instruments, Abingdon, UK) was used to analyze the live image data.
Next-generation sequencing
Total RNA from h-UC-SSEA-3(+) cells and h-BM-Muse cells was isolated by using the NucleoSpin RNA. The extracted RNA's purity and concentration were evaluated using an Agilent 2100 bioanalyzer (Agilent Technologies, Palo Alto, CA, USA). The library was constructed by NEBNext Poly(A) mRNA Magnetic Isolation Module (New England Biolabs [NEB], Hitchin, UK) and NEBNext Ultra RNA Library Prep Kit for Illumina (NEB). Sequencing was performed on a NovaSeq 6000 (Illumina) by Rhelixa (Tokyo, Japan). Quality control and adapter trimming of sequencing data were performed using fastp version 0.20.0[32]. DNA reads were mapped to the human reference genome DNA (build GRCh38.94) by HISAT2 version 2.2.1 with the default setting [33]. Gene counts from the mapped HISAT2 output and gene was generated by featureCounts version 2.0.128 for downstream expression analysis. Read counts for each gene were normalized to transcripts per million.
In vivo dynamics of intravenously injected cells in mouse acute liver damage model
C57BL/6 mice (8–10 weeks old, Japan SLC, Shizuoka, Japan) were habituated under a specific pathogen-free environment at 24 ± 2°C with a 12-h light-dark cycle at least for 1 week and allowed free access to standard institutional food and tap water except during injection. Mice received intraperitoneal injection of 5.0 mL/kg of carbon tetrachloride (CCl4; Fujifilm, Osaka, Japan) on day 0 and day 4. The CCl4 was dissolved in olive oil (Fujifilm) at 1:10. Human-UC-SSEA-3(+) cells (5×104 cells) and h-BM-Muse cells (5×104 cells) were injected into the tail vein of C57BL/6 mice 24 h after the second CCl4 intraperitoneal injection. At 7 days after the cell injection, the mice administered 1 mg AkaLumine-HCl (30 mM, Fujifilm) in distilled water. After 10 min, mice were killed by a lethal dose of isoflurane anesthesia. Each organ was removed, immersed in 500 µM AkaLumine-HCl in normal saline, and evaluated using an IVIS SpectrumCT in vivo imaging system (Perkin Elmer, Waltham, MA, USA). Living Image 4.5 software (Perkin Elmer) was used to analyze the regions of interest. The data are presented as the total photon flux (photons/s).
Fluorescence in situ hybridization for X-inactive specific transcript analysis
The ubiquitously transcribed tetratricopeptide repeat gene on the X chromosome (UTX, known as the X chromosome-specific gene) and X-inactive specific transcript (XIST)-specific probes were used for fluorescence in situ hybridization analysis according to the manufacturer's instructions (Chromosome Science Labo Inc., Sapporo, Japan). Cells were fixed with 4% PFA in 0.1 M PB, treated with 0.005% pepsin (Fujifilm)/0.1 N HCl at 37℃ for 3 min. After washing with PBS, cells were fixed with 4% PFA in 0.1 M PB, and then incubated with UTX-specific and XIST-specific probes overnight at 37℃. The cells were washed with 2x saline sodium citrate (SSC) for 5 min at 37℃ with preheated 50% formamide/2xSSC for 20 min at 37℃ and with 1xSSC for 15 min at room temperature. Samples were counterstained with DAPI (1:500). Images were acquired with a confocal laser microscope (A1; Nikon) and the percent of XIST(+) cells was measured.
Single-cell RNA sequencing (scRNA-seq)
Single-cell capture and cDNA synthesis were performed based on the TAS-seq protocol using the Rhapsody Single-Cell Analysis System (BD) [34]. All libraries were sequenced using Novaseq 6000 (Illumina, San Diego, CA, USA). Adapter trimming of sequencing data was performed using cutadapt version 2.10 [35]. Filtered cell barcode reads were annotated based on the Python script provided by BD. cDNA reads were mapped to reference RNA (build GRCh38 release-101) using bowtie2 version 2.4.2 [36]. The resulting count data were converted to a genes x cells matrix file. The sequencing and sequencing data processing was performed by ImmunoGeneTeqs (Tokyo, Japan).
Seurat R package version 3.2.2 in R version 4.0.2 was used for filtering, data integration, normalization, dimension reduction, and differentially expressed gene detection [37]. Genes detected in fewer than 3 cells were removed at the filtering step. Cells outside the thresholds of > 5000 expressed genes and 1–10% mitochondrial genes were considered low-quality cells and excluded. Our data were integrated with human embryo data: preimplantation stage (E-MTAB-3929) [38], pregastrulation stage (GSE136447) [38], and gastrulation stage (E-MTAB-9388) [39]. Data integration was performed by Seurat's "FindIntegrationAnchors" function with the parameter "dims = 30". After data integration, Seurat's "ScaleData" function performed normalization. Clusters were graphically displayed by uniform manifold approximation and projection (UMAP) plot [40]. Cell clustering was performed by Seurat's "FindNeighbors" and "FindClusters" functions with the parameter "dims = 24, resolution = 0.8". To evaluate the similarity for each cell type, Pearson’s product-moment correlation coefficient was calculated among all cell types using all genes by the "cor" function. We used Seurat's "FindMarkers" function with the MAST algorithm to detect upregulated or downregulated genes of each type of the h-UC-Muse cells and h-embryo cells compared to the remaining cells [41]. Commonly upregulated or downregulated genes in each cell type (cluster 1, 3, 7 in h-UC-Muse cells, as well as epiblast, trophectoderm, primitive endoderm, primitive streak, and advanced mesoderm with a high correlation coefficient with them) were applied to Metascape (http://metascape.org), respectively [42]. Metascape was used for the pathway and process enrichment analysis with default parameters to explore the common biologic processes of h-UC-Muse cells and h-embryo cells [42]. For marker genes of each lineage, we selected genes annotated with Gene Ontology (GO) terms related to lineage development, such as "nervous system development", "heart development", and "liver development". Dot plots and bar plots were generated by ggplot2 R package (https://ggplot2.tidyverse.org.), a heatmap by ComplexHeatmap R package [43, 44], and violin plots by Seurat's "Vlnplot" function.
Enzymatic methyl sequencing
All libraries were prepared using NEBNext Enzymatic Methyl-seq Kit (#E7120; NEB, Ipswich, MA, USA) and converted to libraries for DNBSEQ using MGIEasy Universal Library Conversion Kit (App-A, MGI Tech, Shenzhen, China) and AMPure XP beads (Beckman Coulter, Brea, CA, USA). The libraries were sequenced using DNBSEQ-G400RS (MGI Tech). Adapter trimming of sequencing data was performed using fastp version 0.20.0 [32]. DNA reads were mapped to reference human genome DNA (build hg19.p13), duplicate reads were removed and the methylation call for every single C analyzed was removed, and then the number of methylated and unmethylated reads was counted using bismark version v0.23.1 [45], Bowtie2 version 2.4.5 [46], and samtools version 1.7 [47]. Sequencing was performed by Kazusa DNA Research Institute (Chiba, Japan).
Methylkit R package version 1.18.0 [48] and genomation version 1.24.0 [49] in R version 4.1.1 were used for filtering, determining the methylation level, calculating correlations, and identifying the differentially methylated regions. Our data were integrated with human embryo data (GSE49828) [50]. Genes detected with a coverage of less than 10 reads were removed at the filtering step. To analyze the methylation level of each CpG site, the following algorithm was applied: number of methylated C divided by the total number of methylated C and unmethylated T at the exact positions of the reference genome was calculated as the DNA methylation level of the CpG site. Every CpG site with a read depth ≥ 1 was summed and counted as the total CpG coverage of the sample. For quantifying the DNA methylation level in each sample, only the CpG sites with a coverage of more than 10 reads were considered and subjected to the 100-base pair (bp)-tile-based DNA methylation algorithm [51]. First, the genome was binned into consecutive 100-bp tiles. The number of methylated C divided by the total number of methylated C and unmethylated T captured in the 100-bp tile, was interpreted as the 100-bp tile averaged DNA methylation level. The CpG density of every CpG site was calculated as the total number of all CpG dinucleotides located within 50 bp upstream and 50 bp downstream of this CpG site. In contrast, the CpG density of every 100-bp tile was then calculated as the average CpG density of all CpG sites in this 100-bp tile. Using these data, we generated a correlation matrix, a heatmap with hierarchical clustering, and a principal component analysis (PCA) plot of DNA methylation between human embryos and Muse cells. To explore the common GO terms of embryonic cells and UC-Muse cells in low methylated genes, genes with low methylation (< 10%) were extracted from h-UC-Muse cells with the same level of methylation as h-post-implantation blastocysts (< 1% variation). Genes with a methylation status that varied by more than 30% or less than − 30% in the CpG region around 1000 bp from the transcription start point in h-UC-Muse cells compared to h-BM-, h-AT-, and h-DT-Muse cells, respectively, were extracted. Metascape was used for the pathway and process enrichment analysis with default parameters [42]. For marker genes of each lineage, genes annotated with GO terms related to lineage development, such as triploblastic, extraembryonic-lineage, germline-lineage, and hematopoietic-lineage differentiation, were selected.
In vitro differentiation of Muse cells
In vitro differentiation of Muse cells into extraembryonic-lineage marker (+) cells was conducted according to the previously described method with minor modifications [52]. For extraembryonic-lineage induction, h-UC-, h-BM-, h-AT-, and h-DT-Muse cells (1×104 cells/ cm2) were cultured in DMEM supplemented with 2% FBS, 10 ng/mL bone morphogenic protein-4 (BMP4, Fujifilm), 20 µM SB431542 (Fujifilm), and 20 µM SU5402 (Fujifilm) on adherent culture for 3 weeks.
In vitro differentiation into germline-lineage marker positive cells was conducted according to the previously described method with minor modifications [53]. Human-UC-, h-BM-, h-AT-, and h-DT-Muse cells (1×104 cells/cm2) were cultured in DMEM supplemented with 10% FBS, 50 ng/mL Activin A Fujifilm), 3 µM CHIR99021 (Fujifilm), and 10 µM Y27632 (Fujifilm) on adherent culture for 2 days (incipient mesoderm-like cell induction) [53]. After 2 days of incipient mesoderm-like cell induction, cells were cultured in DMEM supplemented with 10% FBS, 200 ng/mL BMP4, 1000 U/mL LIF (MilliporeSigma), 100 ng/mL stem cell factor (SCF, Fujifilm), 50 ng/mL epidermal growth factor (EGF, Fujifilm), and 10 µM Y27632 (Fujifilm) in suspension culture using low-cell binding V-bottom 96-well plates for 4 days.
In vitro differentiation into hematopoietic-lineage marker (+) cells was conducted according to the previously described method with minor modifications [54]. Human-UC-, h-BM-, h-AT-, and h-DT-Muse cells (1×104 cells/cm2) were cultured in DMEM supplemented with 5% FBS and 100 ng/mL BMP4 on adherent culture for 2 days and then in DMEM supplemented with 5% FBS and 20 ng/mL BMP4 on adherent culture for 2 days. After 4 days of adherent culture, the cells were cultured in DMEM supplemented with 5% FBS, 40 ng/mL vascular endothelial growth factor (Fujifilm), 50 ng/mL SCF for 2 days and then in DMEM supplemented with 5% FBS, 50 ng/mL SCF, 10 ng/mL thrombopoietin (Fujifilm), 50 ng/mL interleukin-3 (Fujifilm), 50 ng/mL granulocyte colony stimulating factor (Fujifilm) and 50 ng/mL FMS-like tyrosine kinase 3 ligand (Fujifilm) in suspension culture using low-cell binding V-bottom 96-well plates for 11 days.
Western blotting
Cells were washed with cold PBS twice and lysed by RIPA Lysis and Extraction Buffer (ThermoFisher Scientific) supplemented with a protease inhibitor (ThermoFisher Scientific) and a phosphatase inhibitor (MilliporeSigma). Cell lysates were centrifuged at 11,000×g at 4℃. The supernatant was transferred to new tubes, and protein amounts were measured using a bicinchoninic acid protein assay kit (ThermoFisher Scientific) according to the manufacturer’s instructions. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) was performed using SDS-PAGE gels, and the proteins were then transferred to polyvinylidene difluoride membranes (MilliporeSigma). Membranes were blocked using 2% BSA/Tris-buffered saline with 0.05% Tween-20 (TBST), then incubated overnight at 4°C in primary antibodies diluted in 2% BSA/TBST. After 3 washes with TBST, the membranes were incubated with secondary antibodies (diluted 1:5000 in 2% BSA/TBST for 1 h at room temperature. After washing with TBST, the blots were developed by Pierce ECL Plus Western Blotting Substrate (ThermoFisher Scientific). Images were acquired using Fusion FX imaging systems (Vilber). Band intensity was quantified by ImageJ software. The primary and secondary antibodies used in this experiment are listed in Supplemental Tables 1 and 2.
Statistics
The data were analyzed by 1-way analysis of variance and an unpaired Student’s t-test to determine statistical significance using Microsoft® Excel software. All data represent 3 independent experiments. The results are presented as the mean ± standard error of the mean. A p-value of less than 0.05 was considered significant.