Journal: Science Advances

Article Title: Pervasive enhanced transcription in inflammatory breast cancer tumors and PBMCs impacts RNA splicing and intronic RNAs in plasma

doi: 10.1126/sciadv.adu0031

Figure Lengend Snippet: Stacked bar graphs showing the percentage of reads in each sample assigned to different RNA biotypes (Ensembl GRCh38 Release 93). ( A ) Major RNA biotype categories excluding rRNA. ( B ) sncRNA categories. ( C ) Repeat-element RNAs categories. ( D ) Different regions of protein-coding genes, including introns, coding sequences (CDSs), and untranslated exon regions (UTRs). ( E ) Protein-coding gene antisense and sense strands. Sample types (FFPE tumors, frozen neighboring healthy breast tissue, frozen non-IBC tumors, PBMCs, or plasma), with disease status (healthy, non-IBC, or IBC), are indicated above. Sample names corresponding to healthy donor or patient ID numbers are indicated below (H, healthy; B, non-IBC; I, IBC). Paired Wilcoxon signed-rank test results for differences in biotype read distributions between sample types can be found in the data file S2.

Article Snippet: Frozen-matched tumor and normal breast tissue RNAs from four deidentified patients with non-IBC were purchased from OriGene ( CR525106 , CR560146 , CR520829 , and CR561033 ).

Techniques: Clinical Proteomics

Journal: Science Advances

Article Title: Pervasive enhanced transcription in inflammatory breast cancer tumors and PBMCs impacts RNA splicing and intronic RNAs in plasma

doi: 10.1126/sciadv.adu0031

Figure Lengend Snippet: ( A ) IDR analysis. Each point indicates the fraction of intron-derived reads for a protein-coding gene in IBC FFPE tumor samples [ y axis, −log 10 (1 − y ) scaling] compared to that of intronic bases in its genomic annotation [ x axis, −log 10 (1 − x ) scaling]. Points are colored by their log 10 -transformed length in base pairs of the genomic interval encoding the gene (exons plus introns): Longer genomic intervals tend to be associated with relatively high genomic intron fractions, evident from the color gradient from left to right. The black line indicates equality y = x , corresponding to the set of points representing genes with IDR = 1; points above or below this line represent genes with IDRs greater or less than 1, respectively. The red trend line plots the solution to the IDR equation (Materials and Methods) when applied to data for all genes, yielding a typical IDR value of ~0.19. ( B ) Upset plots showing the number of genes with IDRs between 0.5 and 1 (top) and IDRs ≥ 1 (bottom) in combined datasets indicated to the right of the plots. ( C ) IGV plots for representative protein-coding genes from combined IBC FFPE tumor datasets. Horizontal arrows for gene maps above the IGV plots indicate the 5′ to 3′ orientation of the gene. Tracks of Ensembl-annotated protein-coding gene exons (short vertical lines) and introns (horizontal lines) and RepBase-annotated Repeat-element RNAs (Rep) are shown on the top. The numbers of reads that mapped to the top (T) or bottom (B) strands are indicated to the right of the IGV plots for each strand. Type I to IV genes are defined in Results and classified into different categories as described in Materials and Methods. Density plots for skewness of intron body coverage are shown as red dashed lines. Asterisks indicate protein-coding genes with IDR > 0.5. Percentages of intron nucleotides corresponding to Repeat-element RNAs are indicated by the arrowhead pointing up to the Rep element track. Percentages of intron reads mapped to Repeat-element RNAs are indicated by the arrowhead pointing down to the IGV plot of mapped reads.

Article Snippet: Frozen-matched tumor and normal breast tissue RNAs from four deidentified patients with non-IBC were purchased from OriGene ( CR525106 , CR560146 , CR520829 , and CR561033 ).

Techniques: Derivative Assay, Transformation Assay

Journal: Science Advances

Article Title: Pervasive enhanced transcription in inflammatory breast cancer tumors and PBMCs impacts RNA splicing and intronic RNAs in plasma

doi: 10.1126/sciadv.adu0031

Figure Lengend Snippet: Tiles are colored by expression difference relative to the gene-wise mean taken across all PBMC samples, as in above. The significantly overexpressed genes shown in the heatmap (adj. P ≤ 0.001 and LFC ≥ 2) are limited to those detected in half or more of the IBC samples. Lanes are labeled at the bottom according to healthy donor and patient ID numbers (table S1). miRNAs mapping to the hairpin miRNA reference were categorized as: m, mature miRNAs, if reads were ≤ 2 nucleotides shorter or longer than the annotated mature miRNA sequence; pre, pre-miRNAs if reads were derived from pre-miRNA only, including relatively long RNAs that were derived from pre-miRNA but shorter than the annotated hairpin sequence; or m + pre, if reads were derived from both mature and pre-miRNAs. Examples of reads for different categories of miRNAs are shown in IGV alignments in fig. S7, D and E. Protein-coding and lncRNA genes with IDR > 0.5 are marked with an asterisk.

Article Snippet: Frozen-matched tumor and normal breast tissue RNAs from four deidentified patients with non-IBC were purchased from OriGene ( CR525106 , CR560146 , CR520829 , and CR561033 ).

Techniques: Expressing, Labeling, Sequencing, Derivative Assay

Journal: Science Advances

Article Title: Pervasive enhanced transcription in inflammatory breast cancer tumors and PBMCs impacts RNA splicing and intronic RNAs in plasma

doi: 10.1126/sciadv.adu0031

Figure Lengend Snippet: ( A to C ) Heatmaps for the indicated sample and RNA comparisons. Tiles are colored by expression differences relative to the corresponding gene-wise means taken across all plasma samples. Lanes are labeled at the bottom according to the healthy donor and patient ID numbers (table S1). miRNAs mapping to the hairpin miRNA reference were categorized as: m, mature miRNAs, if reads were 1 or 2 nucleotides shorter or longer than the annotated mature miRNA sequence. ( D ) IGV plots of overrepresented genes in healthy or non-IBC plasma compared to IBC plasma. T and B, top and bottom DNA strands, respectively. The number of reads for RNAs in each IGV plot is indicated to the right. ( E ) Stacked bar graphs showing the percentage of bases in reads mapped to CDSs, UTRs, introns, or intergenic regions for all protein-coding genes detected in the indicated plasma samples (left three bars) or for significantly overrepresented protein-coding genes (adj. P ≤ 0.001 and LFC ≥ 2) detected in ≥50% of the indicated plasma samples (right four bars).

Article Snippet: Frozen-matched tumor and normal breast tissue RNAs from four deidentified patients with non-IBC were purchased from OriGene ( CR525106 , CR560146 , CR520829 , and CR561033 ).

Techniques: Expressing, Clinical Proteomics, Labeling, Sequencing

Journal: Science Advances

Article Title: Pervasive enhanced transcription in inflammatory breast cancer tumors and PBMCs impacts RNA splicing and intronic RNAs in plasma

doi: 10.1126/sciadv.adu0031

Figure Lengend Snippet: ( A ) Average abundance based on DESeq2 normalized counts for each gene whose RNAs were detected in plasma samples plotted against its average normalized expression level for RNAs in non-IBC or IBC patient FFPE tumors samples (left two columns) or healthy donor, non-IBC patient, or IBC patient PBMC samples (right three columns) for reads mapped to exons + introns (top row), exons only (middle row), or introns only (bottom row). Each point represents a single gene, colored according to its binned IDR values in FFPE tumor or PBMC samples, with color codes shown on the top right plot. Generalized additive model trendlines are fit for each group of IDR-binned genes, with 95% confidence intervals based on SE estimates indicated by the ribbons around the trendlines. The trendlines show separation of gene sets with different IDRs and particularly for the region around 5 on the x axis in each panel, with high-IDR genes (red points and trendlines) tending to show higher average plasma enrichment than lower-IDR genes with similar FFPE tumor or PBMC expression levels. Solid lines represent trends calculated from genes ≤100 kb and dashed lines for genes >100 kb. ( B ) Same comparisons as in (A) with points colored according to selected Hallmark gene sets with color codes shown on the top right plot. Generalized additive model trendlines are fit for each group of IDR-binned genes, with 95% confidence intervals based on SE estimates indicated by the ribbons around the trendlines.

Article Snippet: Frozen-matched tumor and normal breast tissue RNAs from four deidentified patients with non-IBC were purchased from OriGene ( CR525106 , CR560146 , CR520829 , and CR561033 ).

Techniques: Clinical Proteomics, Expressing

Journal: PloS one

Article Title: Reduced selenium-binding protein 1 in breast cancer correlates with poor survival and resistance to the anti-proliferative effects of selenium.

doi: 10.1371/journal.pone.0063702

Figure Lengend Snippet: Figure 1. The Expression of SELENBP1 in Normal and Tumor Breast Tissues. Breast cancer tissue arrays were stained by immunohistochemistry using anti-human SELENBP1 antibody at 1:100 dilution. Positive stained cells are shown in dark brown color. (A) Strong positive staining of SELENBP1 in normal breast tissue under low power view (200X). (B–C) Weak positive to negative staining of SELENBP1 in breast cancer tissues under high power view (400X). (D) The Allred scoring distributions of SELENBP1 expression in normal and tumor tissue groups. Inside lines represent means and standard deviations. *p,0.05. (E) Statistical results for the difference between normal and tumor tissues as analyzed by Kruskal-Wallis test. doi:10.1371/journal.pone.0063702.g001

Article Snippet: Breast Cancer Tissue Array Slides Three sets of breast cancer tissue arrays (including normal, primary tumor, and metastases) were obtained from Imgenex or Biomax Inc.

Techniques: Expressing, Staining, Immunohistochemistry, Negative Staining

Journal: PloS one

Article Title: Reduced selenium-binding protein 1 in breast cancer correlates with poor survival and resistance to the anti-proliferative effects of selenium.

doi: 10.1371/journal.pone.0063702

Figure Lengend Snippet: Figure 2. SELENBP1 Expression is Progressively Reduced in Advancing Clinical Stages in Breast Cancer Tissues. (A) The scoring distributions of SELENBP1 expression in normal tissues and tumor tissues at stage II and stage III. Inside lines represent means and standard deviations. **p,0.01. (B) Statistical results for the difference between normal and tumor tissues as analyzed by Kruskal-Wallis test. (C) Survival curves of breast cancer patients with respect to different SELENBP1 expression levels are shown at stage II and (D) stage III. Blue and red lines represent the SELENBP1-high and SELENBP1-low groups, respectively. doi:10.1371/journal.pone.0063702.g002

Article Snippet: Breast Cancer Tissue Array Slides Three sets of breast cancer tissue arrays (including normal, primary tumor, and metastases) were obtained from Imgenex or Biomax Inc.

Techniques: Expressing

Journal: PloS one

Article Title: Reduced selenium-binding protein 1 in breast cancer correlates with poor survival and resistance to the anti-proliferative effects of selenium.

doi: 10.1371/journal.pone.0063702

Figure Lengend Snippet: Figure 3. The Correlation of SELENBP1 Expression with ER, PR, and TP53 in Breast Cancer Tissues. (A) The scoring distributions of SELENBP1 expression in normal tissues and tumor tissues with ER+ and ER– status. The inside lines represent means and standard deviations. **p,0.01. The difference between normal and ER+ and ER– tumor tissues was analyzed by Kruskal-Wallis test and statistical results are shown (B). Survival curves of breast cancer patients with respect to different SELENBP1 expression are shown in ER+ group in (C). The blue line is the SELENBP1- high group and the red line is the SELENBP1-low group. The scoring distributions of SELENBP1 expression in normal and tumor tissues with PR+/PR–

Article Snippet: Breast Cancer Tissue Array Slides Three sets of breast cancer tissue arrays (including normal, primary tumor, and metastases) were obtained from Imgenex or Biomax Inc.

Techniques: Expressing

Journal: Cancer Science

Article Title: Design and Evaluation of Eb 4 Mab ‐7‐ mG 2a : A Dual‐Action Anti‐ EphB4 Monoclonal Antibody for Targeted Breast Cancer Therapy

doi: 10.1111/cas.70198

Figure Lengend Snippet: Prognostic significance of EphB4 expression in breast cancer and establishment of Eb 4 Mab‐7‐mG 2a . (A) Kaplan–Meier survival analysis of OS in breast cancer patients based on the expression levels of EphB family genes (EphB1, EphB2, EphB3, EphB4, and EphB6) using RNA‐seq data from KMplot. Among these genes, high expression of EphB2 and EphB4 was significantly associated with poor prognosis (log‐rank p < 0.05). Patients were then stratified into high and low expression groups using the optimal cutoff within the interquartile range. (B) Generation of a mouse IgG 2a version of the anti‐EphB4 monoclonal antibody (Eb 4 Mab‐7‐mG 2a ). The V H and V L regions of B4Mab‐7 (mouse IgG 1 ) were cloned into expression vectors containing mouse IgG 2a C H and C L regions. These vectors were transfected into BINDS‐09 cells to produce Eb 4 Mab‐7‐mG 2a . (C) Left: Expression of EphB4 and β‐Actin in CHO‐K1, CHO/EphB4, MCF‐7, and EphB4‐KO MCF‐7 cells. Right: Expression of EphB4 and GAPDH in normal human adult breast tissues, MCF10A (non‐tumorigenic mammary epithelial cells), MCF‐7, BT‐474, SK‐BR‐3, MDA‐MB‐468, and HBC5 cells.

Article Snippet: Aside from cell lysates, a commercially available normal human breast tissue lysate (Total Protein–Human Adult Normal Tissue: Breast, Novus Biologicals, Littleton, CO, USA) was used to assess expression in normal breast tissue.

Techniques: Expressing, RNA Sequencing, Clone Assay, Transfection