Research ArticleMOLECULAR BIOLOGY

S100A8/S100A9 cytokine acts as a transcriptional coactivator during breast cellular transformation

See allHide authors and affiliations

Science Advances  01 Jan 2021:
Vol. 7, no. 1, eabe5357
DOI: 10.1126/sciadv.abe5357
  • Fig. 1 S100A8/A9 is important for breast cellular transformation.

    (A) Ribosome-associated RNA levels in nontransformed [ethanol (EtOH)–treated] and transformed [tamoxifen (TAM)–treated] cells. Differentially expressed genes are colored blue; cytokine genes that are differentially expressed are colored red. The inset shows morphology of cells during TAM-induced transformation. (B) Super-enhancer identification in nontransformed (top) and transformed (bottom) cells, with ranks of enhancers associated with S100A8/A9 shown in the parentheses. Insets show H3K27ac chromatin immunoprecipitation–sequencing (ChIP-seq) signal at the S100A8/A9 genomic region with enhancer or super-enhancer regions marked. (C) Western blot (left) and soft agar assay (right) with transformed cells transfected with negative control (NCi), S100A8, or S100A9 siRNA. Data are means ± SEM; n = 6; ***P < 0.001 compared with negative control treatment; t test. TPR, Translocated Promoter Region, Nuclear Basket Region. (D) Soft agar assay with cells induced by TAM together with indicated antibody treatment. Data are means ± SEM; n = 4; ***P < 0.001 compared with immunoglobulin G (IgG) treatment; t test. (E) RNA expression of S100A8 and S100A9 in PAM50-defined molecular subtypes of breast cancer. Wilcoxon rank sum test P values are shown. A total of 950 clinical breast samples from 840 patients in The Cancer Genome Atlas database were used in this analysis. RPKM, reads per kilobase of transcript per million mapped reads; RPM, reads per million; LDHA, lactate dehydrogenase A; RSEM, RNA-Seq by Expectation-Maximization.

  • Fig. 2 Nuclear S100A8/A9 and breast transformation.

    (A) Western blots of cytosolic (C) and nuclear (N) fractions in nontransformed and transformed cells. LDHA and histone 2B (H2B) are used as subcellular compartment markers. (B) Confocal imaging of S100A8 and S100A9 immunofluorescence costaining in a transformed cell. (C) S100A8 and S100A9 immunohistology from clinical samples of breast cancer (top) and skin cancer (bottom). Representative nuclear localizations of S100A8 or S100A9 are marked by arrows. Scale bar, 50 μm. Data are obtained from The Human Protein Atlas database. (D) Bright-field and GFP fluorescence imaging demonstrate nuclear-specific expression of the indicated derivatives in nontransformed and transformed cells. Scale bar, 50 μm. (E) Western blots of whole-cell lysates show expression of nuclear localization signal (NLS)–GFP–S100A8 (top) and NLS-GFP-S100A9 (bottom). IB, immunoblot. (F) Representative colony formation (top) and quantification (bottom) of soft agar assay with EtOH- or TAM-treated cells expressing indicated protein. Data are means ± SEM; n = 6; ***P < 0.001 compared with corresponding NLS-GFP group; t test. IB, Immunoblot.

  • Fig. 3 Genome-wide chromatin binding of S100A8 and S100A9.

    (A) Western blots of the indicated endogenous and ectopic proteins in whole-cell lysates of EtOH- or TAM-treated cells. (B) Western blots of Flag-tagged S100A8 (top) or S100A9 (bottom) in cytosolic and nuclear fraction under EtOH or TAM treatment. LDHA and snRNP70 are used as subcellular compartment markers. snRNP70, Small Nuclear Ribonucleoprotein U1 subunit 70. (C) Coimmunoprecipitation of Flag-tagged S100A8 with endogenous S100A9 in nuclear extract under EtOH or TAM treatment. HSC70 (Heat Shock Cognate 71 kDa protein) is used as a negative control; LDHA is used as a marker for the cytosolic compartment. (D) Venn diagrams show peak overlap of S100A8 and S100A9 binding in EtOH or TAM treatment. P < 2.2 × 10−16 for any of the peak overlap enrichment; Fisher’s exact test. (E) Relationship of S100A8 and S100A9 binding levels for the union of all peak regions of S100A8 or S100A9. (F) Combined S100A8 and S100A9 binding levels EtOH or TAM conditions. Peaks with statistically significant up- or down-regulation are colored red or blue, respectively. The inset shows the distribution of fold enrichment values for all S100A8/A9 peaks. (G) Binding profiles (average, left) and heatmaps (right) for S100A8 or S100A9 in EtOH- or TAM-treated condition around the summits of the union of all peak regions for S100A8 or S100A9. Replicate-pooled binding data are shown. Heatmaps are sorted by S100A9 ChIP signal in TAM treatment. FDR, false discovery rate; bp, base pairs.

  • Fig. 4 Epigenomic features of S100A8 and S100A9 chromatin binding.

    (A) Chromatin occupancy of indicated factors at ±2000 bp of summits of the union of all peak regions for S100A8 or S100A9 in combination with EtOH- and TAM-treated conditions. Heatmaps are segmented according to types of genomic regions as defined by patterns of histone modifications. Replicate-pooled chromatin binding data are shown for S100A8 and S100A9. Heatmaps are sorted by S100A9 ChIP signal in TAM treatment. (B) Percentage of the indicated classes of regions for all S100A8/A9 sites (A8A9.ALL) and DNase I hypersensitive sites (DNaseHS). (C) Meta-gene profiles (average, left) and heatmaps (right) of Pol II association from 1 kb upstream of transcription start site (TSS) and 2 kb downstream of transcription end site (TES) at indicated groups of S100A8/A9 target genes in EtOH and TAM treatment. Insets show Pol II signal at coding region; dash lines are aligned to 0.15. Heatmaps are sorted by S100A8/A9 ChIP signal density (±250 bp of peak summit) at sites at promoters (Pr), enhancers (Eh), or both (Pr + Eh). (D) Location of S100A8, S100A9, and Pol II signals with respect to the TSS along with box plots indicating the distance between pairwise combinations.

  • Fig. 5 S100A8/A9 binding and cancer gene transcription.

    (A) Box plots on the left show expression levels of genes associated with (coral; 7679 genes) or without (gray; 2627 genes) S100A8/A9 binding sites in EtOH- or TAM-treated cells. Box plot on the right shows the fold change (TAM:EtOH) of the same classes of genes. Higher expression levels in both conditions and higher fold induction during transformation are associated with S100A8/A9-bound sites (Wilcoxon rank sum test P values indicated). (B) Percentage of genes differentially expressed (at least twofold) during transformation that are up-regulated (UP; pink) or down-regulated (DN; blue) and are associated with sites that are bound or unbound by S100A8/A9. Fisher’s exact test P value is indicated for the distribution comparison. (C) Gene counts and percentage of noncancer genes (NonCancer), oncogenes (ONG), TSGs, and genes annotated as both oncogenes and TSGs (OncoTSG) in hg19 refGene collection (refGene), all S100A8/A9 target genes (A8A9.ALL), and genes with strongly up-regulated S100A8/A9 peaks in TAM treatment (A8A9.UP). (D) Gene counts and percentage of cancer (left) and noncancer (right) genes from indicated gene collections. (E) Meta-gene average profiles of Pol II binding from 1 kb upstream of TSS and 2 kb downstream of TES at indicated groups of S100A8/A9 target genes in EtOH- and TAM-treated cells. Gene counts are shown in the parentheses. (F) Induction ratio (fold change relative to the mRNA level in EtOH-treated cells expressing NLS-GFP) of the indicated classes of genes (average of six cancer genes/class or two control genes) in cells expressing the indicated proteins. Data are means ± SEM; ns, not significant; *P < 0.05 compared with corresponding NLS-GFP group; Wilcoxon signed rank test. Asterisks (*) for increase were colored red or otherwise colored blue. RT-qPCR, reverse transcription quantitative polymerase chain reaction.

  • Fig. 6 Motif enrichment and S100A8/A9 chromatin binding.

    (A) De novo identified DNA motifs at S100A8/A9 sites at the indicated classes of genomic regions highly matched to known motifs. The de novo motif search was performed for S100A8/A9 sites against DNase I hypersensitive sites lacking S100A8/A9 association (S100A8/A9-null sites) in the same class of chromatin regions. Hypergeometric P values for motif enrichments are shown. (B) Distributions of de novo discovered motif density at ±200 bp of the indicated peak summits of indicated classes of chromatin regions. (C) TAM-induced fold changes of S100A8/A9 chromatin binding at S100A8/A9 sites with or without indicated motifs (alone or in combination) in promoter regions. *P < 0.05, **P < 1 × 10−5, ***P < 1 × 10−10, and ****P < 2.2 × 10−16 compared with S100A8/A9 sites without indicated motifs (other); Wilcoxon rank sum test. Asterisks (*) for increased fold change were colored red or otherwise colored blue. Peak numbers of indicated groups are shown in the parentheses. The dashed line is aligned to median value of the Other group. (D) Same as (C) for sites at enhancer regions. (E) Same as (C) for sites at repressed regions.

  • Fig. 7 S100A8/A9 interacts with transcription factors and activates transcription upon artificial recruitment.

    (A) Western blot of S100A9-Flag immunoprecipitates of the indicated proteins in nuclear extracts from cells treated with EtOH (E) or TAM (T); lactate dehydrogenase (LDHA) is used as a cytoplasmic marker. (B) Transcriptional activity (normalized luciferase levels) in MCF7 cells electroporated with the indicated Gal4 DNA binding domain (DBD) protein derivative and the reporter construct containing five tandem repeats of the Gal4 binding site upstream of the core Drosophila HSP70Bb TATA-box promoter. Data are means ± SEM; n = 4; ***P < 0.001 compared with the Gal4DBD group; t test.

Supplementary Materials

  • Supplementary Materials

    S100A8/S100A9 cytokine acts as a transcriptional coactivator during breast cellular transformation

    Ruisheng Song, Kevin Struhl

    Download Supplement

    This PDF file includes:

    • Figs. S1 to S6
    • Table S1

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article