Journal: Molecular Cancer

Article Title: AI-based classification of anticancer drugs reveals nucleolar condensation as a predictor of immunogenicity

doi: 10.1186/s12943-024-02189-3

Figure Lengend Snippet: Unbiased morphological profiling of anticancer agents using cell painting assay. ( a ) An overview of the strategy for the screening procedure is illustrated. Human osteosarcoma U2OS cells were treated with a custom-built anticancer library at 3 µM, while cisplatin was used at 300 µM and oxaliplatin at 500 µM, supplemented with a selection of agents inhibiting transcription (i.e. 3 µM epirubicin, 3 µM idarubicin, 10 µM abemaciclib, 1 µM becatecarin, 1 µM trabectedin, 5 µM BMH21, 1 µM CX5461, 15 µM aclarubicin, 1 µM lurbinectedin, 50 µM topotecan, 20 µM metarrestin, 200 µM DRB, 1 µM flavopiridol, 1 µM triptolide) or left untreated (CTR), for 4 h. Subcellular organelles were stained according to the cell painting assay and cells were subjected to high-content microscopy, image segmentation and subsequent cell features extraction. After sub-setting two groups from the dataset, i.e. transcription inhibitors and untreated (CTR) group, a random forest (RF) binary classifier was trained, and feature importances were thereafter extracted from the latter. ( b ) Representative fluorescence and transmitted light (TL)-based images show the panel of acquired channels, i.e. Hoechst 33342 (nucleus, DNA; DAPI channel), MitoTracker Deep Red (mitochondria; Cy5 channel), concanavalin A (endoplasmic reticulum; FITC channel), SYTO 14 (nucleoli, cytoplasmic RNA; Cy3 channel), phalloidin (actin; Texas Red channel) and TL (brightfield), for a selection of agents. Scale bar equals 10 μm. ( c ) The displayed relative Gini index decrease represents the importance of features for classification. Colors represent the acquisition channels from which the feature was extracted. ( d ) Hierarchical cluster dendrogram based on the weighted distance matrix calculated using Gower’s similarity coefficient identifies a grouping pattern of the evaluated chemical compounds. Agents from the custom-built anticancer library are depicted in blue, supplemental transcription inhibitors in rose red, anticancer transcription inhibitors in brown and an adjacent green dot indicates confirmed immunogenic cell death (ICD) inducers

Article Snippet: Fig. 4 Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1μM, 10μM and 100μM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU).

Techniques: Selection, Staining, Microscopy, Extraction, Fluorescence

Journal: Molecular Cancer

Article Title: AI-based classification of anticancer drugs reveals nucleolar condensation as a predictor of immunogenicity

doi: 10.1186/s12943-024-02189-3

Figure Lengend Snippet: Unboxing deep neural networks discloses condensation of nucleoli (CON) as utmost relevant subcellular phenotypic feature. ( a ) Human osteosarcoma U2OS cells were treated with a panel of chemical agents, i.e. 3 µM etoposide (ETO), 3 µM mitoxantrone (MTX), 3 µM doxorubicin (DOXO), 500 µM oxaliplatin (OXA), 3 µM dactinomycin (DACT), 15 µM aclarubicin (ACLA), 1 µM BMH21, 5 µM plicamycin (PLICA), or left untreated (CTR) for 4 h, imaged using transmitted light (TL) and representative images are displayed. After automated nuclei detection using semantic segmentation, patches were extracted, subsampled and then classified using trained Deep Convolutional Neural Network (DCNN). Gradient-weighted Class Activation Mapping (Grad-CAM) technique was used to map principal features from the final convolutional layer to the original image. The heatmap generated from Grad-CAM highlighting the crucial regions for the network’s prediction is overlayed to the original TL image. Color represents the degree of activation (importance) from low (0, black), through medium (0.5, red), to high (1, yellow). The DCNN-predicted probability values (p) of condensation of nucleoli (CON) are superimposed on each image. ( b ) The initial DNN classifier was generalized by enriching the training set with additional micrographs of new cell types (e.g. H9C2) and various treatments (including cytotoxic agents). Left panel: Top numbers indicate the newly trained classes (0: controls; 1: early transcription inhibitors; 2: late cytotoxic drugs), exemplified by representative images from selected conditions. Right panel: the training process is depicted as the evolution of accuracy (plain dots) and loss (circles) along epochs. Blue and red colors represent the training and validation sets, respectively. ( c ) Manually selected regions of interest (ROI) from TL micrographs, generated in the Cell Painting assay (Fig. ) after treatment with DACT or untreated (CTR), were resized and used to generate GradCAM images with the trained DCNN. These images were then overlayed to both original TL and SYTO 14-stained ROIs. The predicted probabilities (p) for CON induction are superimposed on each TL image. ( d ) Two generative adversarial networks (GANs) were trained to generate images of CON − and CON + classes. Examples of image series belonging to the two classes synthesized by GANs are depicted over multiple epochs during the training process, ranging from 0 to 10,000. ( e ) Boxplots represent the CON probabilities calculated from the DCNN of a panel of images belonging to synthetic CON - and CON + class generated by GANs. ( f ) U2OS cells were left untreated (CTR), or treated with 10 µM ETO, 3 µM MTX, 3 µM DOXO, 500 µM OXA, 1 µM DACT, 5 µM ACLA, 1 µM BMH21 or 5 µM PLICA and subsequently imaged via time-lapse holotomographic microscopy every 20 min for a total duration of ~ 16 h. Representative holotomographic images together with the segmentation overlay of the nucleus (in white) and of nucleoli (in different colors) are displayed. ( g ) Kinetics of total areas of nucleoli over that of nuclei are displayed as median ± MAD. ( h ) Nucleoli total area over nuclear area was extracted for 4 h and 16 h timepoints and represented in bar charts as median ± MAD with dots representing each segmented cell. P-values were calculated using pairwise Mann-Whitney test against adequate control. Scale bars represent 10 μm

Article Snippet: Fig. 4 Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1μM, 10μM and 100μM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU).

Techniques: Activation Assay, Generated, Biomarker Discovery, Staining, Synthesized, Microscopy, MANN-WHITNEY, Control

Journal: Molecular Cancer

Article Title: AI-based classification of anticancer drugs reveals nucleolar condensation as a predictor of immunogenicity

doi: 10.1186/s12943-024-02189-3

Figure Lengend Snippet: Aclarubicin interferes with RNA synthesis, does not cause DNA damage and is immunogenic in cultured cells and in murine vaccination.( a ) Human osteosarcoma U2OS cells were treated with 5 µM aclarubicin (ACLA), 1 µM dactinomycin (DACT), 3 µM mitoxantrone (MTX), or left untreated (CTR) for 4 h. After fixation, nuclear counterstaining with Hoechst 33342 and permeabilization, cells were further stained with anti-Phospho-Histone H2A.X (Ser 139) antibody, images were acquired by fluorescence microscopy and analyzed. Nuclear γH2AX intensity was quantified, and representative images are displayed. ( b ) U2OS cells were pre-treated as previously listed for 2.5 h and treatment pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU). After fixation and nuclear counterstaining with Hoechst 33342, cells were permeabilized, EU was stained with an Alexa Fluor-488-coupled azide, images were acquired by fluorescence microscopy and analyzed. Nuclear EU intensity was quantified, and representative images are displayed. ( c ) H2AX phosphorylation (γH2AX) and EU incorporation were assessed as described above. To quantify the co-occurrence of fibrillarin (FBL) and nucleolin (NCL), U2OS cells were treated as previously listed for 4 h before fixation and permeabilization. Cells were then stained with the combination of a rabbit anti-fibrillarin and a mouse anti-nucleolin antibodies, followed by a staining with anti-rabbit Alexa Fluor-564-coupled and anti-mouse Alexa Fluor-488-coupled antibody, respectively. The CON phenotype was assessed using a trained Convolutional Neural Network (CNN) after automatically extracting nuclei images from TL micrographs through semantic segmentation. The effect of H2AX phosphorylation, EU incorporation, co-occurrence of FBL/NCL and CON phenotype from transmitted light (TL) was ranged between 0 and 100% of the effect of the positive control mitoxantrone (MTX) and values are summarized in an heatmap. ( d ) U2OS wild-type (WT) cells were treated with a dose range of ACLA (5 µM, 10 µM and 20 µM), with 2 µM MTX or 500 µM oxaliplatin (OXA) as positive controls or left untreated (CTR) for 24 h. The concentration of ATP secreted in the supernatant was quantified with a luciferase-based bioluminescence kit. ( e ) U2OS WT cells were treated as described above and the concentrations of HMGB1 released in the supernatant was quantified with a specific ELISA kit. ( f ) U2OS WT cells were treated for 6 h, medium was refreshed for additional 24 h and after collection, the surface-exposed calreticulin (CALR) was stained with a specific antibody. DAPI was used as an exclusion dye, cells were acquired by flow cytometry and the percentage of CALR + cells among viable (DAPI − ) ones is depicted. ( g ) Histograms representing CALR intensity are illustrated for each condition and percentages of CALR + DAPI − are reported. ( h ) U2OS WT cells were treated for 6 h and medium was refreshed for supplemental 24 h. Supernatants were transferred on HT29 MX1-GFP reporter cells for 48 h, and 100 ng/mL human type 1a interferon (IFNα1) was added as an additional positive control. Representative images are displayed. ( i ) The percentage of MX1 + cells is reported in a bar chart. ( j ) U2OS WT and U2OS cells expressing non-phosphorylatable eIF2αS51A (S51A) were treated for 6 h with previously listed treatments, with the addition of 3 µM thapsigargin (THAPS) as positive control. The phosphorylation of eIF2α (PeIF2α) was assessed by immunofluorescence staining with a phosphoneoepitope-specific eIF2α antibody and representative images are displayed. ( k ) PeIF2α cytoplasmic intensity was quantified, and represented as a bar chart. ( l ) Experimental outline of prophylactic vaccination. Mouse fibrosarcoma MCA205 cells were treated in vitro with 15 µM ACLA or 4 µM MTX. Dying cells were harvested and subcutaneously ( s.c. ) injected into the left flank of immunocompetent syngeneic C57Bl/6 mice ( n = 10 mice per group), while the control group was injected with PBS. Ten days later, animals were rechallenged with living MCA205 cells in the contralateral flank of the mice and tumor size was regularly monitored. ( m ) The course of tumor volume curves is depicted. ( n ) Tumor-free survival is shown for each group. ( o ) ACLA- and MTX-vaccinated tumor free mice (ACLA n = 10, MTX n = 10) as well as naïve mice ( n = 5) were rechallenged by s.c. injection of living MCA205 cells in the right flank of the mice and tumor growth was monitored. The dot plot indicates tumor volumes at endpoint for each group. Values in bar charts are expressed as mean ± SD of one representative out of three independent experiments. Significant p-values for in vitro assays were determined by pairwise Mann-Whitney test versus relevant control. Values in tumor growth curves are expressed as mean ± SEM, and median of tumor sizes is reported for the endpoint dot plot with p-values calculated with pairwise Mann-Whitney test versus control. TumGrowth ( https://github.com/kroemerlab ) was used to analyze in vivo data. Statistical significance of tumor-free survival was calculated with log-rank test. Scale bars equal 10 μm

Article Snippet: Fig. 4 Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1μM, 10μM and 100μM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU).

Techniques: Cell Culture, Staining, Fluorescence, Microscopy, Phospho-proteomics, Positive Control, Concentration Assay, Luciferase, Enzyme-linked Immunosorbent Assay, Flow Cytometry, Expressing, Immunofluorescence, In Vitro, Injection, Control, MANN-WHITNEY, In Vivo

Journal: Molecular Cancer

Article Title: AI-based classification of anticancer drugs reveals nucleolar condensation as a predictor of immunogenicity

doi: 10.1186/s12943-024-02189-3

Figure Lengend Snippet: Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1µM, 10µM and 100µM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU). After fixation and nuclear counterstaining with Hoechst 33342, cells were permeabilized, EU was stained with an Alexa Fluor-488-coupled azide and cells were further stained with anti-Phospho-Histone H2A.X (Ser 139) antibody (γH2AX), followed by staining with Alexa Fluor 647 secondary antibody. To assess cell death U2OS cells were treated for 24 h, fixed and counterstained with Hoechst 33342. Images were acquired by fluorescence and transmitted light (TL) microscopy. Representative images of cells treated with 5 µM aclarubicin (ACLA), 10 µM etoposide (ETO), 300 µM cisplatin (CDDP), 1 µM dactinomycin (DACT), 500 µM oxaliplatin (OXA) or left untreated (CTR) are displayed. Scale bar equals 10 μm. A colormap representing confluency of cells is depicted for each condition. Scale bar represents 100 μm. ( b ) The clustered heatmap summarizes the fraction of maximum effects of each assessed parameter for 4 groups of interest (i.e. “ACLA-like”, “ETO-like”, cisplatin “CDDP-like” and doxorubicin “DOXO-like”) at a chosen concentration that maximized the overall effect of the assessed parameters at 4 h. Cell death was quantified via enumeration of Hoechst-stained nuclei, DNA damage activity via nuclear H2AX phosphorylation (γH2AX), inhibition of transcription via EU intensity in the nucleus and condensation of nucleoli (CON) was evaluated by a trained convolutional neural network (CNN) applied to nuclei patches from TL images. Color-codes indicating DNA interaction (DNA int.) type and enzymatic target (Enzym.) are reported for each condition. TOPO, topoisomerase; UNS, unspecific; POL, polymerase; LIG, ligase; GYR, gyrase. Adjacent green dots highlight confirmed immunogenic cell death (ICD) inducers while rose red dots indicates hits chosen for further validations. Data of each group are also reported in boxplots, CON in ( c ), EU in ( d ), γH2AX in ( e ) and cell death in ( f ). P-values were calculated using a pairwise Mann-Whitney test against the “DOXO-LIKE” group for each assessed parameter

Article Snippet: Fig. 4 Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1μM, 10μM and 100μM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU).

Techniques: Staining, Fluorescence, Microscopy, Concentration Assay, Activity Assay, Phospho-proteomics, Inhibition, MANN-WHITNEY

Journal: Molecular Cancer

Article Title: AI-based classification of anticancer drugs reveals nucleolar condensation as a predictor of immunogenicity

doi: 10.1186/s12943-024-02189-3

Figure Lengend Snippet: Non-DNA-damaging agents are effective in prophylactic vaccination, which mostly correlates with condensation of nucleoli (CON) and cell death. ( a ) Human osteosarcoma U2OS cells expressing photoactivatable (PA)GFP-H2A were used to evaluate hits’ chromatin-damaging activity after activation by laser of a portion of the nucleus. Segmentation of nuclei in colors (each color represents each nucleus that has been automatically tracked over time) and photoactivated region (PR) in blue are overlayed on representative cells treated with 10 µM doxorubicin (DOXO) or left untreated (CTR). Photoactivated PAGFP-H2A was subsequently monitored by time-lapse confocal microscopy for 60 min and tracked over time. ( b ) Partial nuclear photoactivation was performed as described above, cells were then treated with 10 µM etoposide (ETO), 10 µM aclarubicin (ACLA), 10 µM bisantrene (BISA), 10 µM BMH21, 10 µM CX5461, 5 µM plicamycin (PLICA), 10 µM doxorubicin (DOXO), or left untreated (CTR). Representative images of the first (0 min) and last (60 min) timepoint are displayed, with a segmentation overlay of nuclei in colors (each color represents each nucleus that has been tracked over time) and in the PR in blue. ( c ) Quantification of the fluorescence intensity of PAGFP-H2A in the PR expressed as fold change (FC) to the initial timepoint (T0) after each indicated treatment is reported over time. ( d ) Quantification of the fluorescence intensity of PAGFP-H2A expressed as FC to T0 in the PR after each treatment is reported in a bar chart corresponding to the last timepoint (60 min). Scale bars equal 10 µm. ( e ) Mouse fibrosarcoma MCA205 cells were treated in vitro with 4 µM mitoxantrone (MTX), 15 µM aclarubicin (ACLA), 5 µM BMH21, 15 µM CX5461, 10 µM bisantrene (BISA) or 50 µM plicamycin (PLICA). After harvesting, dying cells were subcutaneous ( s.c. ) injected into the left flank of immunocompetent syngeneic C57Bl/6 mice ( n = 10 mice per group), while the control group was injected with PBS. Ten days later, animals were rechallenged with living MCA205 cells in the contralateral flank of the mice and tumor size was regularly measured. The course of tumor volume curves is depicted. ( f ) Tumor-free survival is shown for each group. ( g ) Vaccinated tumor free mice (ACLA n = 9, BMH21 n = 5, CX5461 n = 10, BISA n = 5, PLICA n = 4, MTX n = 9) as well as naïve mice ( n = 5) were rechallenged by s.c. injection of living MCA205 cells in the left flank of the mice, with simultaneously s.c. injection of living mouse melanoma B16-F10 in the right flank. Tumor growth was monitored over time and tumor volumes at endpoint of each group are displayed in a dot plot for B16-F10, and in ( h ) for MCA205. Values in tumor growth curves are expressed as mean ± SEM, and median of tumor sizes are reported for the endpoint dot plot with p-values calculated with pairwise Mann-Whitney test versus control. TumGrowth ( https://github.com/kroemerlab ) was used to analyze in vivo data. Statistical significance of tumor-free survival was calculated with log-rank test. ( i ) The clustered heatmap summarizes the relative effect of each evaluated phenotype, i.e. vaccination efficacy (Vacc. efficacy), condensation of nucleoli (CON), death, nucleolar size, chromatin damage (Chro. Damage), phosphorylation of eIF2α (PeIF2α), dsDNA and γH2AX, assessed throughout the study for ACLA, BMH21, BISA, CX5461, PLICA, etoposide (ETO), oxaliplatin (OXA), dactinomycin (DACT), doxorubicin (DOXO) and the untreated (CTR) group. ( j ) A Pearson correlation matrix for the aforementioned effects is displayed. Pearson correlation coefficients are indicated in the lower triangular matrix, while being represented by circles whose size and color are mapped to their value in the upper triangular matrix

Article Snippet: Fig. 4 Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1μM, 10μM and 100μM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU).

Techniques: Expressing, Activity Assay, Activation Assay, Confocal Microscopy, Fluorescence, In Vitro, Injection, Control, MANN-WHITNEY, In Vivo, Phospho-proteomics

Journal: Molecular Cancer

Article Title: AI-based classification of anticancer drugs reveals nucleolar condensation as a predictor of immunogenicity

doi: 10.1186/s12943-024-02189-3

Figure Lengend Snippet: Prediction of condensation of nucleoli (CON) through Quantitative Structure-Activity Relationship (QSAR) modelling using neural network and molecular descriptors. ( a ) The experimental outline for training set preparation is displayed. Human osteosarcoma U2OS cells were treated with the different libraries for 4 h. After fixation, images were acquired using transmitted light (TL) microscopy and extracted nuclei were classified by a convolutional neural network in condensation of nucleoli (CON) negative (CON − ) or positive (CON + ) phenotype. After a filtration step performed on the whole dataset based on Tanimoto similarity for duplicates removal, training and validation pies specify the number and corresponding percentage of CON − and CON + . ( b ) The libraries used in the screening campaign are presented in a table, each row specifying the library’s name as described above, its number of compounds (N) and the percentage of compounds classified as CON + . Libraries are further divided into training set in black and validation (VAL.) set in red. ( c ) The QSAR procedure is represented in a flowchart. Chemistry Development Kit (CDK) library was used for computing 116 descriptors of the compounds, some of which exemplified in a table. These descriptors were thereafter used to define an applicability domain and then fed to 3 different algorithms, i.e. natural network (NN), random forest and xgboost, for establishing a model capable to predict the CON phenotype. ( d ) Performance parameters statistics from the Monte Carlo cross-validation (MCCV) procedure of the different algorithms are reported. ( e ) Validation of the neural network model using y-scrambling procedure. The same NN model architecture was trained with 100 randomly shuffled training labels, and both loss and AUC were evaluated and reported in a bi-parametric plot. Dots represent the median for each group (black = y-scrambles; red = models from the initial cross-validation (CV) procedure). Dotted lined represent point densities. P-values calculated by means of the Mann-Whitney test against the MCCV original trainings’ values are shown. ( f ) The ~ 320,000 compounds of the NCI database were annotated retrieving MeSH (Medical Subject Headings) terms. The predicted CON score was calculated with the neural network-based QSAR model previously trained. Cumulative distribution is plotted in black for the full data set ( n = 319,481) and in red for data associated with MeSH terms ( n = 10,948). ( g ) Compounds associated with MeSH records were split into two categories, according to their propensity to induce the CON phenotype, and the occurrence of each term (expressed as the fraction of compounds associated with it) was computed. The occurrence ratio and -log(p-value) from a χ 2 test between the two populations for each MeSH term are displayed in a biparametric graph

Article Snippet: Fig. 4 Classification of drugs subtypes based on nuclear and toxicity features from a screening campaign. ( a ) Human osteosarcoma U2OS cells were pre-treated with a custom-made collection of 274 DNA/RNA-related drugs from TargetMol at 1μM, 10μM and 100μM or left untreated (CTR) for 2.5 h. Treatments pursued for 1.5 additional hours in the presence of 1mM 5-ethynyl uridine (EU).

Techniques: Activity Assay, Microscopy, Filtration, Biomarker Discovery, MANN-WHITNEY

Journal: Oncology reports

Article Title: miRNA-regulated expression of oncogenes and tumor suppressor genes in the cisplatin-inhibited growth of K562 cells.

doi: 10.3892/or_00000813

Figure Lengend Snippet: Figure 2. Changes in mRNA expression of BCL2, E2F1, E2F3, RB1 and P53 in K562 cells after cisplatin treatment. BCL2, E2F1, E2F3, RB1 and P53 were detected by (A) RT-PCR, (B) real-time PCR and (C) ELISA.

Article Snippet: K562 cells were lysed [lysis buffer: 0.15 M NaCl, 5 mM EDTA (pH 8.0), 1% Triton X-100, 10 mM Tris-Cl (pH 7.4), 100 mM PMSF and 5 M DTT) and incubated in a 96-well plate, followed by the addition of goat anti-human antibodies against BCL2 (1:400), E2F1, E2F3 (1:400, 1:400), RB1 (1:500) and P53 (1:300, 1:400) (all from Immunoleader, Boster) according to the manufacturer's instructions.

Techniques: Expressing, Reverse Transcription Polymerase Chain Reaction, Real-time Polymerase Chain Reaction, Enzyme-linked Immunosorbent Assay

Journal: Oncology reports

Article Title: miRNA-regulated expression of oncogenes and tumor suppressor genes in the cisplatin-inhibited growth of K562 cells.

doi: 10.3892/or_00000813

Figure Lengend Snippet: Figure 4. Correlative expression of miRNAs and oncogenes using antisense oligos (ASO). (A) RT-PCR, (B) real-time PCR. Correlative expression of (C) E2F1 and its targeted miR-17-5p, (D) E2F3 and its targeted miRNAs and (E) Bcl-2 and its targeted miRNAs (miR-16, 34a-c) using ELISA is shown. *Significant difference (p<0.05).

Article Snippet: K562 cells were lysed [lysis buffer: 0.15 M NaCl, 5 mM EDTA (pH 8.0), 1% Triton X-100, 10 mM Tris-Cl (pH 7.4), 100 mM PMSF and 5 M DTT) and incubated in a 96-well plate, followed by the addition of goat anti-human antibodies against BCL2 (1:400), E2F1, E2F3 (1:400, 1:400), RB1 (1:500) and P53 (1:300, 1:400) (all from Immunoleader, Boster) according to the manufacturer's instructions.

Techniques: Expressing, Reverse Transcription Polymerase Chain Reaction, Real-time Polymerase Chain Reaction, Enzyme-linked Immunosorbent Assay

Journal: Oncology reports

Article Title: miRNA-regulated expression of oncogenes and tumor suppressor genes in the cisplatin-inhibited growth of K562 cells.

doi: 10.3892/or_00000813

Figure Lengend Snippet: Figure 5. Expression of E2F1, E2F3, BCL2, RB1 and P53 genes regulated by miRNAs. Detection of the expression of (A) E2F1, (B) E2F3, (C) BCL2, (D) RB1 and (E) P53. *Significant difference (p<0.05).

Article Snippet: K562 cells were lysed [lysis buffer: 0.15 M NaCl, 5 mM EDTA (pH 8.0), 1% Triton X-100, 10 mM Tris-Cl (pH 7.4), 100 mM PMSF and 5 M DTT) and incubated in a 96-well plate, followed by the addition of goat anti-human antibodies against BCL2 (1:400), E2F1, E2F3 (1:400, 1:400), RB1 (1:500) and P53 (1:300, 1:400) (all from Immunoleader, Boster) according to the manufacturer's instructions.

Techniques: Expressing

Journal: The Korean Journal of Physiology & Pharmacology : Official Journal of the Korean Physiological Society and the Korean Society of Pharmacology

Article Title: Murrayafoline A Induces a G 0 /G 1 -Phase Arrest in Platelet-Derived Growth Factor-Stimulated Vascular Smooth Muscle Cells

doi: 10.4196/kjpp.2015.19.5.421

Figure Lengend Snippet: Effects of murrayafoline A on the inhibition of cell cycle regulatory proteins. Quiescent VSMCs cultured in serum-free medium were stimulated with PDGF-BB to express cell cycle regulatory proteins, and the effects of murrayafoline A on the expression of cyclin E, CDK2, cyclin D1, CDK4, and PCNA, and activation of pRb were assessed as described in the Experimental Section. β-Actin was used for normalization. Immunoblots were analyzed by densitometry and the values are given based on the control of 1.0. The results are an average of four similar experiments, expressed as means±SEM. The insets display representative blots of four similar independent experiments. Statistical differences from the PDGF-BB control (PDGF-BB-stimulated, but no murrayafoline A) are indicated by * p<0.05 or ** p<0.01.

Article Snippet: Anti-phospho-pRb, anti-CDK2, anti-CDK4, anti-phospho PCNA, anti-cyclin D1, anti-cyclin E, anti-Akt, and anti-β-actin antibodies were purchased from Abfrontier (Geumcheon, Seoul, Korea).

Techniques: Inhibition, Cell Culture, Expressing, Activation Assay, Western Blot