Journal: International Journal of Medical Sciences

Article Title: Controlling Osteogenesis and Adipogenesis of Mesenchymal Stromal Cells by Regulating A Circadian Clock Protein with Laser Irradiation

doi:

Figure Lengend Snippet: Intracellular location of (A) mCRY1 and (B) mPER2 proteins in MSCs 24 hr after laser irradiation. Cells were double-labeled with DAPI (blue, upper panel) and mCRY1 or mPER2 (red, center panel). The lower panel provides a merged image. mCRY1 and mPER2 localized to the cytoplasm prior to laser irradiation. After laser irradiation, proteins translocated to the nucleus. (C) mRNA levels of mCry1 in MSCs 24 hr after laser irradiation (100 mW/cm ) and in non-irradiated cells. Samples were normalized to mRps18 . The mRNA levels of mCry1 decreased after blue laser irradiation relative to non-irradiated cells. Scale bars = 30µm *, p<0.01: significant difference between the relative mRNA levels of laser irradiated MSCs and controls.

Article Snippet: The mouse MSCs cell line (KUSA-A) was purchased from RIKEN Bioresource Center, Japan, and cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal calf serum (FCS), 100 units/mL penicillin, and 0.1 mg/mL streptomycin at 37°C in a 5% CO 2 atmosphere.

Techniques: Irradiation, Labeling

Journal: International Journal of Medical Sciences

Article Title: Controlling Osteogenesis and Adipogenesis of Mesenchymal Stromal Cells by Regulating A Circadian Clock Protein with Laser Irradiation

doi:

Figure Lengend Snippet: (A) The beam profile of the blue laser (wave length; 405 nm, continuous wave). MSCs were irradiated for 180 sec at various intensities. (B) Histochemical analysis of laser irradiated MSCs. Alizarin red-S staining of irradiated MSCs (magnification: x50). At 5 days post-irradiation, calcium deposition had increased around the cells in a dose-dependent manner. Calcium phosphate deposition was evaluated by von Kossa staining (magnification: x50). At 5 days after treatment, staining increased with increased laser energy. The area expressing alkaline phosphatase (ALP) activity was stained (magnification: x50). Laser irradiated samples displayed immunopositive staining for osteocalcin, a marker of osteoblast differentiation (magnification: x100). Scale bars = 200 (for Alizarin red-S, von Kossa, and ALA staining) and 100µm (for osteocalcin immunostaining). (C) The quantitative calcium content increased after blue laser irradiation relative to non-irradiated cells. Calcium content increases varied with laser energy. *, p<0.01: significant difference between the calcium content of laser-irradiated MSCs and controls.

Article Snippet: The mouse MSCs cell line (KUSA-A) was purchased from RIKEN Bioresource Center, Japan, and cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal calf serum (FCS), 100 units/mL penicillin, and 0.1 mg/mL streptomycin at 37°C in a 5% CO 2 atmosphere.

Techniques: Irradiation, Staining, Expressing, Activity Assay, Marker, Immunostaining

Journal: International Journal of Medical Sciences

Article Title: Controlling Osteogenesis and Adipogenesis of Mesenchymal Stromal Cells by Regulating A Circadian Clock Protein with Laser Irradiation

doi:

Figure Lengend Snippet: (A) Staining with oil red O demonstrates that blue laser irradiation decreased adipogenesis relative to non-irradiated areas (magnification: x50). Higher magnification (x400) is shown in frame. Scale bars = 200 µm. As the nuclear localization of CRY proteins attenuates CLOCK- and BMAL1-driven transcription, laser irradiation may act 'molecular switch' for regulatory proteins by suppressing CRY transcription to limit the accumulation of lipid droplets in cells. (B) mRNA levels of PPARγ in MSCs 24 hr after laser irradiation (100 mW/cm 2 ) and in non-irradiated cells. Samples were normalized to mRps18 . The mRNA levels of PPARγ decreased after blue laser irradiation relative to non-irradiated cells. *, p<0.01: significant difference between the relative mRNA levels of laser irradiated MSCs and controls.

Article Snippet: The mouse MSCs cell line (KUSA-A) was purchased from RIKEN Bioresource Center, Japan, and cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal calf serum (FCS), 100 units/mL penicillin, and 0.1 mg/mL streptomycin at 37°C in a 5% CO 2 atmosphere.

Techniques: Staining, Irradiation

Journal: PLoS ONE

Article Title: Transcription factors, coregulators, and epigenetic marks are linearly correlated and highly redundant

doi: 10.1371/journal.pone.0186324

Figure Lengend Snippet: ( A ) Histogram of Pearson’s r values between measured and predicted values using 10-fold CV for all marks around the TSSs of protein coding genes in K562 cells. ( B,C,D ) Scatter plot comparing predicted and measured values (10-fold CV) for ( B ) DNA methylation, ( C ) H3K4me3, and ( D ) H3K27me3 around the TSSs of protein coding genes in K562 cells. The line “ y = x ” is indicated in red for reference. ( E ) Mark weight distribution in the linear model fitted for CEBPB on 100% of the data around TSSs of protein coding genes in K562. ( F ) Barplot of selected mark types for different mark types from the linear models fitted for all marks on 100% of the data for TSSs of protein coding genes in H1. We considered the four different mark types (chromatin remodelers, coregulators, epigenetic marks, and transcription factors) and calculated the relative frequency of each mark type (dark blue bars). Then, for each mark we considered all mark weights of these four types, i.e., without those with an unknown respectively not regulating function. We took a 95% quantile cutoff over all absolute weights, where we considered the weights for all mark models combined. For each mark type, we considered those weights in the linear models for each mark of that respective type, whose absolute weight was above the cutoff, grouped these weights according to the type of the input mark, and plotted the respective relative frequency of each input mark type in the bars of the same color. The bars, where the predicted mark type and the input mark type are identical, are marked with a red star.

Article Snippet: Furthermore, in the NIH Roadmap Epigenomics Consortium the largest datasets for histone modifications and DNA methylation data were available for the cell lines H1, H9, and IMR90.

Techniques: DNA Methylation Assay

Journal: PLoS ONE

Article Title: Transcription factors, coregulators, and epigenetic marks are linearly correlated and highly redundant

doi: 10.1371/journal.pone.0186324

Figure Lengend Snippet: ( A ) Scatter plot for the Pearson’s r comparison for each mark (with data for at least two cell lines) between the median correlation between predicted and measured values, when the models, with which the predictions are made, are fitted in other cell lines (on all marks that are present in both cell lines), and the intracellular models. For the first part for each respective mark (with data for at least two cell lines), for each locus constellation, and all ordered pairs of different cell lines, where data for the mark of interest is available for both cell lines, we fit a model for that mark on all other epigenetic marks, for which data is available for both cell lines, in the first cell line, predict the enrichment of epigenetic mark of interest in the second cell line, calculate the Pearson’s r between predicted and measured values, and take the median over all these values for that mark. For the second part for each respective mark (with data for at least two cell lines), we take the median over all 10-fold CV Pearson’s r values of the intracellular models between predicted and measured data for that mark over each locus constellation and cell line, where there is data for that mark available. The median over these median values is shown as dashed red lines. The solid line “ y = x ” is indicated in red for reference. ( B ) Barplot of median Pearson’s r in the cross-cell line setting and the 10-fold CV setting, where we displayed just those marks with a decrease of at least 0.3. Marks labeled in red are known to have a silencing function and the marks labeled in green are positively associated with gene expression. ( C ) Boxplot of difference between median Pearson’s r in the cross-cell line setting and the 10-fold CV setting for those marks with a decrease of at least 0.3 (left panel) and activating histone modifications, which means here all histone acetylations and H3K4 methylations (right panel). (Two-sided two-sample Kolmogorov-Smirnov test: D = 1, P = 1.132e-10) ( D,E ) Scatter plots between predicted and measured values for ( D ) H3K27me3 and ( E ) H3K4me3 for TSSs of protein coding genes in GM12878 cells, when the model was fitted in H1. The line “ y = x ” is indicated in red in the scatter plots for reference. ( F ) Heatmap showing median Pearson’s r between predicted and measured values for TSSs of protein coding genes over all marks in that target cell line, that are also present in the starting cell line, where the models for the prediction are fitted in the starting cell line and then used to predict the enrichments in the target cell line. For each entry, the target cell line is named as the row entry and the starting cell line is named as the column entry.

Article Snippet: Furthermore, in the NIH Roadmap Epigenomics Consortium the largest datasets for histone modifications and DNA methylation data were available for the cell lines H1, H9, and IMR90.

Techniques: Comparison, Labeling, Gene Expression

Journal: PLoS ONE

Article Title: Transcription factors, coregulators, and epigenetic marks are linearly correlated and highly redundant

doi: 10.1371/journal.pone.0186324

Figure Lengend Snippet: ( A ) Scatter plot between predicted and measured values (when using 10-fold CV) for CAGE gene expression for protein coding genes in K562 cells when 40-bin resolution data was taken for the input marks of the MARS model (pseudocount ε optimized). ( B ) Barplot of Pearson’s r (when using 10-fold CV) for different models for protein coding genes. The bar labels are encoded by their model index, where the first letter represents the cell line (K = K562, G = GM12878, H = H1, I = IMR90), the middle symbols stands for the data input (1 = 1-bin resolution, 40 = 40-bin resolution, 40 m = middle two bins for each mark in 40-bin resolution), and the latter represents the model type (L = linear model, M = MARS model). For each of these the pseudocount ε was optimized. ( C ) Scatter plot between predicted and measured values for CAGE gene expression for the protein coding genes in GM12878, when a MARS model on 40-bin resolution data was fitted in K562 cells. The pseudocount ε is the same for calculating the logarithmized gene expression in both cell lines by using the optimized ε for K562 cells in the 10-fold CV setting. ( D ) Barplot of Pearson’s r values for protein coding genes, when considering each possible ordered pair of different cell lines (analogous to ( C ), with labels as in ( B )), shown in blue, and the Pearson’s r (when using 10-fold CV) for individual cell lines, shown in red, when using MARS models with 40-bin resolution.

Article Snippet: Furthermore, in the NIH Roadmap Epigenomics Consortium the largest datasets for histone modifications and DNA methylation data were available for the cell lines H1, H9, and IMR90.

Techniques: Gene Expression

Journal: iScience

Article Title: Charting oncogenicity of genes and variants across lineages via multiplexed screens in teratomas

doi: 10.1016/j.isci.2021.103149

Figure Lengend Snippet: Identification of significantly enriched drivers and cell types (A) Growth kinetics of round 1 teratoma formation for injections with driver library transduced hESCs vs control WT hESCs. (B) UMAP visualization of cell types from round 1 teratomas formed by driver library transduced hESCs. (C) Growth kinetics of round 2 tumors formed from re-injected cells from round 1 teratomas formed by driver library transduced hESCs vs WT hESCs. Control measurements are from a common set of tumors grown from the parent WT hESC line, which were used as growth controls for all experiments in this study. (D) UMAP visualization of cell types from round 2 tumors formed by re-injected cells from round 1 teratomas of driver library transduced hESCs. (E) Relative fraction of each cell type in round 1 teratomas formed from library transduced hESCs and WT hESCs, and log fold change with associated -log(p-value) of each cell type for driver library teratomas vs WT teratomas. (F) Relative fraction of each cell type in round 2 tumors formed from library transduced hESCs and WT hESCs, and log fold change with associated -log(p-value) of each cell type for driver library tumors vs WT tumors. (G) Relative fraction of top enriched drivers prior to injection and in each round of tumor formation. (H) H&E stained sections of round 2 tumors formed from WT and driver library transduced hESCs. WT tumors display mature cell types from all three germ layers, such as cartilage (top right), muscle (bottom right) and dermis-like epithelium (bottom left). Driver library tumors display disorganized and more homogeneous composition along with markers of transformation such as nuclear pleomorphism (top right), areas of high mitotic rates (bottom right) and areas of necrosis (bottom left). Scale bars for full sections are 5 mm, scale bars for magnified images are 50 μm.

Article Snippet: H1 male hESC cell line was maintained under feeder-free conditions in mTeSR1 medium (Stem Cell Technologies) supplemented with 1% antibiotic-antimycotic (Thermo Fisher Scientific).

Techniques: Control, Injection, Staining, Transformation Assay

Journal: iScience

Article Title: Charting oncogenicity of genes and variants across lineages via multiplexed screens in teratomas

doi: 10.1016/j.isci.2021.103149

Figure Lengend Snippet: Identification of significantly enriched drivers and cell types for tumors formed with driver libraries without c-MYC and myr-AKT1 (A) Growth kinetics of round 1 teratoma formation for injections with driver library transduced hESCs vs WT hESCs. (B) UMAP visualization of cell types from round 1 teratomas formed by driver library transduced hESCs. (C) Growth kinetics of round 2 tumors formed from re-injected cells from round 1 teratomas formed by driver library transduced hESCs vs WT hESCs. Control measurements are from the common set of tumors grown from the parent WT hESC line, which were used as growth controls for all experiments in this study. (D) UMAP visualization of cell types from round 2 tumors formed by re-injected cells from round 1 teratomas of driver library transduced hESCs. (E) Relative fraction of each cell type in round 1 teratomas formed from library transduced hESCs and WT hESCs, and log fold change with associated -log(p-value) of each cell type for driver library teratomas vs WT teratomas. (F) Relative fraction of each cell type in round 2 tumors formed from library transduced hESCs and WT hESCs, and log fold change with associated -log(p-value) of each cell type for driver library tumors vs WT tumors. (G) Relative fraction of top enriched drivers prior to injection and in each round of tumor formation. (H) H&E stained sections of round 1 and round 2 tumors formed from hESCs transduced with driver libraries without c-MYC or myr-AKT1 . Round 1 tumors contain neuroectodermal and epithelial cell types as the majority of cells, while round 2 tumors contain mesenchymal cell types as the majority of cells. Scale bars for full sections are 5 mm, scale bars for magnified images are 100 μm.

Article Snippet: H1 male hESC cell line was maintained under feeder-free conditions in mTeSR1 medium (Stem Cell Technologies) supplemented with 1% antibiotic-antimycotic (Thermo Fisher Scientific).

Techniques: Injection, Control, Staining, Transduction

Journal: iScience

Article Title: Charting oncogenicity of genes and variants across lineages via multiplexed screens in teratomas

doi: 10.1016/j.isci.2021.103149

Figure Lengend Snippet: Validation of tumor formation and proliferative advantages of significantly enriched drivers from library screens (A) Growth kinetics of round 1 and round 2 tumors formed from a mixture of hESCs transduced with driver hits (c-MYC, myr-AKT1, c-MYC + myr-AKT1 or MEK1 (S218D, S222D) and hESCs transduced with a negative control (mCherry). Control measurements are from the common set of tumors grown from the parent WT hESC line, which were used as growth controls for all experiments in this study. (B) Fraction of reads detecting either the driver or negative control barcodes at each stage: pre-injection, round 1 tumor formation and round 2 tumor formation. Barcodes were amplified from genomic DNA. (C) Gene expression of lineage-specific markers for round 2 tumors driven by individual hits. Expression values are normalized and log transformed. (D) H&E stain of round 2 tumor driven by myr-AKT1 showing regions of poor differentiation (top right) and necrosis (bottom right) interspersed with regions of organized tissue (bottom left). (E) H&E stain of round 2 tumor driven by c-MYC showing regions of poor differentiation (top and bottom right) and regions of necrosis (bottom left). (F) H&E stain of round 2 tumor driven by c-MYC + myr-AKT1 showing regions of poor differentiation (top and bottom right) and regions of necrosis (bottom left). (G) H&E stain of round 2 tumor driven by MEK1 (S218D, S222D) showing primarily regions of mature mesenchymal fibroblast-like tissue (top and bottom right) and cartilage-like (bottom left). (D–G) Scale bars for full sections are 5 mm, scale bars for magnified images are 50 μm. (H) Immunofluorescence micrographs of DAPI and Ki-67 stained sections. Scale bars are 75 μm.

Article Snippet: H1 male hESC cell line was maintained under feeder-free conditions in mTeSR1 medium (Stem Cell Technologies) supplemented with 1% antibiotic-antimycotic (Thermo Fisher Scientific).

Techniques: Biomarker Discovery, Transduction, Negative Control, Control, Injection, Amplification, Gene Expression, Expressing, Transformation Assay, Staining, Immunofluorescence

Journal: The Journal of Experimental Medicine

Article Title: Aberrant ZNF423 impedes B cell differentiation and is linked to adverse outcome of ETV6-RUNX1 negative B precursor acute lymphoblastic leukemia

doi: 10.1084/jem.20130497

Figure Lengend Snippet: ZNF423 expression is regulated by DNA methylation. (A) Transactivation of ZNF423 isoform-specific promoters in dependence of central CGI. Reporter gene assay using ZNF423 promoters (α and β) with or without CGI after transfection in 293T cells. Firefly luciferase activity was normalized to Renilla luciferase activity. Error bars represent SD from three technical replicates. Significance is calculated by Student’s t test, comparing indicated samples (***, P ≤ 0.001). Data were reproduced in three independent experiments. (B) Methylation map of CpGs in the central CGI at the ZNF423 locus. Genomic DNA from primary ALL ( n = 58), matched BM MNCs in complete continuous remission (MNC [CCR]; n = 58), H1, HES2 ESC lines, and normal hematopoietic cells at various stages of differentiation ( n = 42) from healthy donors were sequenced after bisulfite conversion. Each row represents one cytosine in a CG dinucleotide of the analyzed sequence. For cell lineage abbreviations, refer to . Significance is calculated by Student’s t test, comparing primary ALL to immunologically characterized control samples (in brackets; *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001). Color code represents degree of methylation in percent. White boxes indicate no sequencing data available. (C) ZNF423 promoter activity in dependence of CPG mutational status. Disruption of CpGs was performed by site-directed mutagenesis as indicated by red marks. Combined deletion mutant (pCpGL-CGI-del_comb_α-prom) represents deletions at CGI positions 102, 183, 220, 250. Wild-type and mutated plasmids were treated with DNA-methylase M.SssI and transfected into 293T. Firefly luciferase activity was normalized to Renilla luciferase activity. Error bars represent SD from three technical replicates. Significance is calculated by Student’s t test, comparing M.SssI-treated wild-type and mutated samples (***, P ≤ 0.001). Data were reproduced in three independent experiments. (D) DNA methylation pattern of central CGI. Diagram was created with BiQ Analyzer software from Max-Planck-Institute of Informatics. Drawing is to scale. White lollipop, unmethylated; black lollipop, methylated. Marked CpGs refer to heatmap in . (E and F) Expression of ZNF423 transcripts upon DNA demethylation and BMP2 stimulation. SEM cells (E) and CD34 + cord blood cells (F) were treated with 3 µM 5A2D for 24, 48, and 72 h. In SEM cells, additional BMP2 treatment was performed 6 h before lysis. Relative fold induction was measured by qPCR (2 −ΔΔCt ) using a ZNF423 isoform-specific primer design in SEM cells. mRNA levels were normalized to B2M and solvent control. Error bars represent SD out of three technical replicates. Significance is calculated using 2 −ΔCt values by Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; n.s., not significant). Data were reproduced in three independent experiments.

Article Snippet: CD133 + fetal liver cells were obtained from StemCell Technologies, Inc. RNA and genomic DNA from embryonic stem cell lines H1 and HES2 were isolated from cell culture leftover materials in accordance with national law.

Techniques: Expressing, DNA Methylation Assay, Reporter Gene Assay, Transfection, Luciferase, Activity Assay, Methylation, Sequencing, Control, Disruption, Mutagenesis, Software, Lysis, Solvent

Journal: Cell stem cell

Article Title: Large-Scale Profiling Reveals the Influence of Genetic Variation on Gene Expression in Human Induced Pluripotent Stem Cells

doi: 10.1016/j.stem.2017.03.009

Figure Lengend Snippet: eQTL Functional Annotation Enrichments. -log10 Fisher exact enrichment p-values for 4,616 eQTL lead SNVs/indels in (A) Roadmap Epigenomics DNase hypersensitivity sites (DHSs) and (B) ENCODE DHSs. The replicate H1 hESC DHS experiments in (B) were performed in different laboratories which may account for their different levels of enrichment. (C) Fisher exact odds ratios for ENCODE H1 hESC transcription factor CHiP-seq peaks. Color indicates whether the enrichment was significant which can vary due to the number of ChIP-seq peaks for each particular mark. See also Table S3.

Article Snippet: We obtained DHSs for the H1, H9, iPS DF 6.9, and iPS DF 19.11 cell lines from Roadmap Epigenomics and merged them into one bed file.

Techniques: Functional Assay, ChIP-sequencing

Journal: Cell stem cell

Article Title: Large-Scale Profiling Reveals the Influence of Genetic Variation on Gene Expression in Human Induced Pluripotent Stem Cells

doi: 10.1016/j.stem.2017.03.009

Figure Lengend Snippet: peQTN Characteristics. (A) Number of stem cell DHSs overlapped by eQTNs for four stem cell lines from Roadmap Epigenomics (H1, H9, iPS DF 6.9, iPS DF 19.11). (B) Number of peQTNs per eGene for 1,526 eGenes with at least one peQTN. (C) Putative eQTN for MED30 that overlaps a CEBPB ChIP-seq peak and disrupts a known CEBPB motif. Scatter plot shows -log10 association p-value from EMMAX for peQTN (purple point), other significant eQTL variants (green points) and variants not associated with MED30 expression (blue points). ENCODE H1 hESC CEBPB ChIP-seq peaks (blue rectangles) and chromHMM chromatin state predictions (multi-color track) are displayed. (D) CTCF ChIP-seq peak coverage (z-score of log10 counts) from five iPSC lines for peaks containing peQTNs predicted to disrupt CTCF binding and that had evidence of CTCF allelic bias in the ChIP-seq data. Counts are stratified by the genotype of the peQTN: 0, 1, and 2 for low, intermediate, and high predicted binding of CTCF, respectively. The peak coverage is significantly associated with the peQTN genotype (r=0.087, p=0.039). (E) The P1 region with the reference allele drives expression in 96% of embryos while the alternate allele leads to a complete loss of expression. (F and G) Image showing tailbud stage Ciona embryo electroporated with (F) Snail P1 (ref. allele) > GFP or (G) Snail P1 (alt. allele) > GFP. Expression can be seen in the tail muscle for the reference allele. Images were taken at the same exposure time to allow for direct comparison. See also Figure S2 and Tables S4 and S5.

Article Snippet: We obtained DHSs for the H1, H9, iPS DF 6.9, and iPS DF 19.11 cell lines from Roadmap Epigenomics and merged them into one bed file.

Techniques: ChIP-sequencing, Expressing, Binding Assay, Comparison