Periostin Exon-21 Antibody Neutralization of Triple-Negative Breast Cancer Cell-Derived Periostin Regulates Tumor-Associated Macrophage Polarization and Angiogenesis

Figures

• 

  Figure 1. Identification of Pn-secreting SUM159PT cells. (A) Construction of fluorescence-based alternative splicing reporter minigenes for Pn exon 21. A genomic fragment from intron 20 to intron 21 of the POSTN gene is inserted into an artificial intron in EGFP cDNA. mCherry cDNA is connected downstream of the EGFP in another reading frame. EGFP and mCherry are expressed when POSTN exon 21 is skipped and included, respectively. (B) Images of SUM159PT cells with the Pn-splicing reporter. (C) Flow cytometry analysis of GFP and mCherry expression on SUM159PT tumor cells. SUM159PT cells with (right) or without (left) the Pn-splicing reporter. Double positivity indicates the fraction positive for both GFP and mCherry. (D) Pn mRNA expression in the cells from the GFP- or double-positive fraction. Pn21 and Pn20/22 indicate endogenous Pn mRNAs with exon 21 and without exon 21, respectively. n = 10. (E) Pn protein expression in conditioned media or cell lysates from GFP- or double-positive cells. Pn was detected with an antibody recognizing exon 12. Three bands indicate ASVs of Pn. Loading condition was confirmed with CBB staining. (F) Immunoprecipitation (IP) studies for Pn exon 21 in conditioned…

• 

  Figure 2. RNA-seq analysis comparing GFP- and double-positive SUM159PT. (A) RNA-seq analysis showing a heat map of differentially expressed genes (DEGs) in GFP- and double-positive SUM159PT. The data represent three biological replicates. Among 1890 DEGs, 1350 were upregulated in double-positive cells, while 540 were upregulated in GFP-positive cells. (B) Gene ontology enrichment of double-positive cells (DP) and GFP-positive cells (GFP). The lengths of the bars indicate the -log 10-transformed p-values. Each dot represents the proportion of genes with the GO term among DEGs. (C) KEGG analysis of upregulated genes in the double-positive cells (DP) and the GFP-positive cells (GFP) from the RNA-seq data. Three biological repeats of each experiment were conducted. (D) Upregulated genes in double-positive cells (DP) include several genes related to macrophage differentiation and recruitment. Data are the log2 of fold change (LogFC). Relative expression pattern analysis of upregulated genes in DP by qRT-PCR analysis to validate the RNA-seq data is shown in Figure S7.

• 

  Figure 3. Pn ASV-specific antibody (Pn-21 Ab) inhibited the growth of double-positive SUM 159PT xenografts. (A) In total, 5 × 106 cells from SUM159PT GFP- and double-positive cells were separately injected into the abdominal mammary glands of athymic nude mice. (B) Images of tumor development at day 7 and 16 post injection. While GFP-positive cells did not form tumors (0/6), double-positive cells engrafted and developed tumors (6/6). (C) Athymic nude mice bearing SUM159PT double-positive cell tumors were treated with Pn-21 Ab or vehicle (n = 7 for Pn-21 Ab, n = 6 for CTRL). The vehicle group received saline, and the other group was treated with Pn-21 Ab (10 mg/kg) 2 times per week throughout the experiments starting at day 0. Tumor volumes were recorded as mean ± SD. * p < 0.05. (D) Pn-21 Ab reduced proliferation in SUM159PT double-positive cell xenografts. Paraffin-embedded tumor sections from the CTRL and Pn-21 Ab groups were stained with antibodies against Ki67 and human nuclei. Ki67-positive human nuclei were quantified. Representative photomicrographs are presented, and quantitative data of Ki67-positive human nuclei are shown in box plots. The p value was calculated using the…

• 

  Figure 4. Disrupting Pn ASV with Pn-21 Ab reduces M2 tumor-supportive TAMs in vivo. Immunohistochemical or immunofluorescent staining with the pan macrophage marker (Iba1) (A), M1 macrophage marker (CD11c) (B), and M2 macrophage markers (CD163 and CD206) (C) in SUM159PT double-positive cell xenografts. Tumors were collected at day 36. Scale bars represent 50 μm. Box plot showing a significant reduction in M2 TAM density in Pn-21 Ab treated xenografts (n = 6–7). The TAM marker-positive area of tumors was quantitated in at least 4 view fields per tumor by using Image J program. *** p < 0.05 vs. CTRL, IgG, respectively. p values were calculated by using the one-way ANOVA test and Tukey–Kramer’s post hoc test.

• 

  Figure 5. Pn with exon 21 reduces M1 macrophage markers but enhances M2 macrophage markers. U937 M0 macrophages were treated with recombinant full-length Pn protein, Pn-1. (A,B) RT-qPCR analyses showing upregulation of M2 TAM markers induced by recombinant Pn-1 in THP-1 (A) or U937 (B) macrophages. PMA-treated macrophages were stimulated with recombinant Pn (500 ng/mL) in RPMI media supplemented with 0.1% BSA for 24 h. RT-qPCR analysis indicated downregulation of M1 markers (iNOS, IL-1β and IL-12a) and up-regulation of M2 markers (CD206, CD163 and Fizz1). Data are shown in mean ± SD, n = 4 per group, * p < 0.05 vs. BSA. Student’s t-test. (C,D) Immunohistochemical staining of CD31 (C) or vWF (D), an endothelial cell marker. CD31- or vWF-positive area in tumors was quantitated in at least 4 view fields per tumor by using Image J program. (n = 6–7). *, ** p < 0.05 vs. CTRL, IgG, respectively. Box plot showing mean % of capillary area in tumors. Scale bars represent 100μm. p values were calculated by using the one-way ANOVA test and Tukey–Kramer’s post hoc test.