Limits...
A Progesterone-CXCR4 Axis Controls Mammary Progenitor Cell Fate in the Adult Gland

View Article: PubMed Central - PubMed

ABSTRACT

Progesterone drives mammary stem and progenitor cell dynamics through paracrine mechanisms that are currently not well understood. Here, we demonstrate that CXCR4, the receptor for stromal-derived factor 1 (SDF-1; CXC12), is a crucial instructor of hormone-induced mammary stem and progenitor cell function. Progesterone elicits specific changes in the transcriptome of basal and luminal mammary epithelial populations, where CXCL12 and CXCR4 represent a putative ligand-receptor pair. In situ, CXCL12 localizes to progesterone-receptor-positive luminal cells, whereas CXCR4 is induced in both basal and luminal compartments in a progesterone-dependent manner. Pharmacological inhibition of CXCR4 signaling abrogates progesterone-directed expansion of basal (CD24+CD49fhi) and luminal (CD24+CD49flo) subsets. This is accompanied by a marked reduction in CD49b+SCA-1− luminal progenitors, their functional capacity, and lobuloalveologenesis. These findings uncover CXCL12 and CXCR4 as novel paracrine effectors of hormone signaling in the adult mammary gland, and present a new avenue for potentially targeting progenitor cell growth and malignant transformation in breast cancer.

No MeSH data available.


Related in: MedlinePlus

Computational Analyses of Microarray Expression Profiles of Mammary Subsets and mRNA Level Validation(A) Schema of experimental pipeline.(B) Heatmaps display levels of mRNA expression with a variance of >3 across all 40 samples.(C) Volcano plots of differential mRNA abundance levels following hormone treatments in distinct cell compartments. X axis: coefficients from linear model in log2 scale; y axis: p values adjusted for multiple testing. The red dotted line depicts q value < 0.05.(D) Venn diagram of significant genes with q value < 0.05 and /fold-change/ ≥ 1.5.(E) Dotmap of significant progesterone-responsive pathways in different cell populations as determined by GOMiner analyses. FDR values are represented by the shade of the gray boxes, and log2 enrichment scores are indicated by the size of the circle within each box.(F) Independent qPCR validation of FACS-sorted basal and luminal samples. Mean ± SEM, EP n = 4, E n = 3 mice/group. Significance levels based on two-tailed unpaired t test, ∗p < 0.05.See also Figure S1.
© Copyright Policy - CC BY-NC-ND
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4376056&req=5

fig1: Computational Analyses of Microarray Expression Profiles of Mammary Subsets and mRNA Level Validation(A) Schema of experimental pipeline.(B) Heatmaps display levels of mRNA expression with a variance of >3 across all 40 samples.(C) Volcano plots of differential mRNA abundance levels following hormone treatments in distinct cell compartments. X axis: coefficients from linear model in log2 scale; y axis: p values adjusted for multiple testing. The red dotted line depicts q value < 0.05.(D) Venn diagram of significant genes with q value < 0.05 and /fold-change/ ≥ 1.5.(E) Dotmap of significant progesterone-responsive pathways in different cell populations as determined by GOMiner analyses. FDR values are represented by the shade of the gray boxes, and log2 enrichment scores are indicated by the size of the circle within each box.(F) Independent qPCR validation of FACS-sorted basal and luminal samples. Mean ± SEM, EP n = 4, E n = 3 mice/group. Significance levels based on two-tailed unpaired t test, ∗p < 0.05.See also Figure S1.

Mentions: Mammary stem cells within the basal epithelium and CD49b+SCA-1− luminal progenitors are both hormone-receptor negative (HR−) (Asselin-Labat et al., 2006; Shehata et al., 2012), and we hypothesized that these cells respond to ovarian hormones via paracrine signals provided by HR+ luminal cells. To delineate the full spectrum of hormone-induced changes in the adult mammary gland, we generated microarray expression profiles of fluorescence-activated cell sorting (FACS)-purified basal (CD24−CD49fhi), luminal (CD24+CD49flo), and stromal (CD24−CD49f−) mammary cells from mice (Figure 1A) under defined hormone treatments (progesterone [P], 17β-estradiol [E], and 17β-estradiol plus progesterone [EP]) or vehicle controls. Unsupervised hierarchical clustering revealed a strong concordance within each cellular compartment, with tighter clustering among EP-treated samples compared with E, P, and vehicle treatments (Figure 1B). Visualization of hormone effects via volcano plots and Venn diagrams indicated that there were more differentially expressed genes in luminal and basal cells in the EP samples than in either the E or P samples alone, compared with vehicle (p < 2.2 × 10−16; proportions test; Figures 1C, 1D, S1A, and S1B). These data are tabulated in Figure S1C. Notably, treatment of mice with P alone had a limited effect on gene expression (Figures 1C and S1A–S1C). This is likely because estrogen is required to drive expression of the progesterone receptor (PR) in adult mice. Venn diagrams also illustrate greater overlap between basal and luminal cells after EP treatment versus E or P alone (Figures 1D and S1A). These data suggest that a combination of E and P generates more robust cellular responses, and these changes occur primarily in the epithelial subsets rather than in the stromal fraction.


A Progesterone-CXCR4 Axis Controls Mammary Progenitor Cell Fate in the Adult Gland
Computational Analyses of Microarray Expression Profiles of Mammary Subsets and mRNA Level Validation(A) Schema of experimental pipeline.(B) Heatmaps display levels of mRNA expression with a variance of >3 across all 40 samples.(C) Volcano plots of differential mRNA abundance levels following hormone treatments in distinct cell compartments. X axis: coefficients from linear model in log2 scale; y axis: p values adjusted for multiple testing. The red dotted line depicts q value < 0.05.(D) Venn diagram of significant genes with q value < 0.05 and /fold-change/ ≥ 1.5.(E) Dotmap of significant progesterone-responsive pathways in different cell populations as determined by GOMiner analyses. FDR values are represented by the shade of the gray boxes, and log2 enrichment scores are indicated by the size of the circle within each box.(F) Independent qPCR validation of FACS-sorted basal and luminal samples. Mean ± SEM, EP n = 4, E n = 3 mice/group. Significance levels based on two-tailed unpaired t test, ∗p < 0.05.See also Figure S1.
© Copyright Policy - CC BY-NC-ND
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4376056&req=5

fig1: Computational Analyses of Microarray Expression Profiles of Mammary Subsets and mRNA Level Validation(A) Schema of experimental pipeline.(B) Heatmaps display levels of mRNA expression with a variance of >3 across all 40 samples.(C) Volcano plots of differential mRNA abundance levels following hormone treatments in distinct cell compartments. X axis: coefficients from linear model in log2 scale; y axis: p values adjusted for multiple testing. The red dotted line depicts q value < 0.05.(D) Venn diagram of significant genes with q value < 0.05 and /fold-change/ ≥ 1.5.(E) Dotmap of significant progesterone-responsive pathways in different cell populations as determined by GOMiner analyses. FDR values are represented by the shade of the gray boxes, and log2 enrichment scores are indicated by the size of the circle within each box.(F) Independent qPCR validation of FACS-sorted basal and luminal samples. Mean ± SEM, EP n = 4, E n = 3 mice/group. Significance levels based on two-tailed unpaired t test, ∗p < 0.05.See also Figure S1.
Mentions: Mammary stem cells within the basal epithelium and CD49b+SCA-1− luminal progenitors are both hormone-receptor negative (HR−) (Asselin-Labat et al., 2006; Shehata et al., 2012), and we hypothesized that these cells respond to ovarian hormones via paracrine signals provided by HR+ luminal cells. To delineate the full spectrum of hormone-induced changes in the adult mammary gland, we generated microarray expression profiles of fluorescence-activated cell sorting (FACS)-purified basal (CD24−CD49fhi), luminal (CD24+CD49flo), and stromal (CD24−CD49f−) mammary cells from mice (Figure 1A) under defined hormone treatments (progesterone [P], 17β-estradiol [E], and 17β-estradiol plus progesterone [EP]) or vehicle controls. Unsupervised hierarchical clustering revealed a strong concordance within each cellular compartment, with tighter clustering among EP-treated samples compared with E, P, and vehicle treatments (Figure 1B). Visualization of hormone effects via volcano plots and Venn diagrams indicated that there were more differentially expressed genes in luminal and basal cells in the EP samples than in either the E or P samples alone, compared with vehicle (p < 2.2 × 10−16; proportions test; Figures 1C, 1D, S1A, and S1B). These data are tabulated in Figure S1C. Notably, treatment of mice with P alone had a limited effect on gene expression (Figures 1C and S1A–S1C). This is likely because estrogen is required to drive expression of the progesterone receptor (PR) in adult mice. Venn diagrams also illustrate greater overlap between basal and luminal cells after EP treatment versus E or P alone (Figures 1D and S1A). These data suggest that a combination of E and P generates more robust cellular responses, and these changes occur primarily in the epithelial subsets rather than in the stromal fraction.

View Article: PubMed Central - PubMed

ABSTRACT

Progesterone drives mammary stem and progenitor cell dynamics through paracrine mechanisms that are currently not well understood. Here, we demonstrate that CXCR4, the receptor for stromal-derived factor 1 (SDF-1; CXC12), is a crucial instructor of hormone-induced mammary stem and progenitor cell function. Progesterone elicits specific changes in the transcriptome of basal and luminal mammary epithelial populations, where CXCL12 and CXCR4 represent a putative ligand-receptor pair. In situ, CXCL12 localizes to progesterone-receptor-positive luminal cells, whereas CXCR4 is induced in both basal and luminal compartments in a progesterone-dependent manner. Pharmacological inhibition of CXCR4 signaling abrogates progesterone-directed expansion of basal (CD24+CD49fhi) and luminal (CD24+CD49flo) subsets. This is accompanied by a marked reduction in CD49b+SCA-1&minus; luminal progenitors, their functional capacity, and lobuloalveologenesis. These findings uncover CXCL12 and CXCR4 as novel paracrine effectors of hormone signaling in the adult mammary gland, and present a new avenue for potentially targeting progenitor cell growth and malignant transformation in breast cancer.

No MeSH data available.


Related in: MedlinePlus