Table 1.
Primer sequences for real time RT-PCR.
Figure 1.
ATBF1 is significantly induced during the differentiation of MCF10A cells in a 3-D culture model.
(A–C) Immunofluorescence staining was used to detect the expression of differentiation markers milk protein β-casein (lower row) (green) (A) and CK18 (red) (B) as well as ATBF1 (green) (C). DAPI staining (blue) shows nuclei of cells. (D–F) Relative mRNA expression of ATBF1, CK18, CK5, and CK14 in MCF10A cells from 2-D and 3-D cultures of indicated times (D and E) or 2-D culture of different times (F), as detected by real time RT-PCR. GAPDH was used as an internal control. Error bar represents SEM. * indicates P<0.01 (student's t test).
Figure 2.
Knockdown of ATBF1 interrupts mammosphere formation and reduces the expression of basal cell markers.
(A, B) Cells transfected with negative control siRNA (NCsi) or ATBF1 siRNA (ATBF1si) were subjected to real time RT-PCR for mRNA expression (A) and western blotting for protein expression (B). (C) Distribution of diameters of all mammospheres at 2-weeks of culture in Matrigel. The average sphere size for ATBF1 knockdown is significantly smaller than that for the control. (D) Knockdown of ATBF1 significantly reduces the number of spheres with a diameter >75 µm. The data are presented as the number of defined mammospheres per 1,000 seeded cells ± SEM. (E) Representative bright field images for spheres from both groups. (F) Measurement of cell numbers for MCF10A cells treated with NCsi or ATBF1si in 24-well plates (2-D culture). (G) Expression of both basal cell markers (CD44, CK14 and CK5) and luminal cell markers (CD24, CK18 and CK8) in MCF10A cells in Matrigel, as detected by real time RT-PCR. * and ** indicate P<0.05 and P<0.01 respectively (student's t test).
Figure 3.
Validation of mammary epithelial cell-specific deletion of Atbf1 in mice.
(A) Increased expression of Atbf1 mRNA in mammary glands of C57BL/6J female mice during puberty and the first week of lactation, as measured by real time RT-PCR. Three mice were used for each time point. Expression of ER, which changes at different stages, was used as a positive control. P values for comparisons between puberty and lactation, between pregnancy and lactation, and between puberty and pregnancy are <0.005, <0.005, and 0.962, respectively. (B) Double IF staining of Atbf1 (red) and SMA (green) shows the expression and localization of Atbf1 in mouse mammary gland. (C) Deletion of Atbf1 genomic DNA upon the expression of Cre in mammary tissues, as detected by PCR. The upper panel shows the wild-type Atbf1 allele (Wt), floxed allele without deletion (Floxed), and the allele with Cre-mediated deletion (Deleted). The lower panel shows the presence of the Cre gene in a mouse, with the interleukin-2 gene (Il-2) as a PCR control. Genotypes and Cre status are indicated at the top. (D) Detection of truncated Atbf1 mRNA in mouse tissues expressing the Cre gene under the MMTV promoter. All mice are positive for MMTV-Cre. “+” and “−” indicate wild-type (Wt) and deleted Atbf1 mRNA respectively. Each lane represents one mouse and two mice were used for each genotype. (E, F) Reduced Atbf1 protein expression by Cre-mediated deletion, as detected by western blotting (E) and immunohistochemical staining (F) with antibody against ATBF1. Thoracic mammary glands from mice with wild-type (+/+), heterozygous (+/−) and homozygous (−/−) Atbf1 were used in these analyses.
Figure 4.
Deletion of Atbf1 promotes ductal elongation and bifurcation in pubertal mammary gland.
(A) Hematoxylin-stained whole-mount images of the fourth abdominal mammary glands from mice at indicated ages. Panels at far right are higher magnification images of indicted areas from 8-week old mammary glands. Three to five mice were used for each genotype. (B–F) Determination of degree of ductal invasion and the numbers of branches and TEBs in mammary glands of mice with indicated genotypes and ages of 5 or 6 weeks (B–D) or 8 weeks (E, F). Degree of ductal invasion was determined by dividing the duct length by the mammary gland length from mid-point of lymph node; and the numbers of total branches and TEBs were determined in whole-mount images by the ImageJ program. (G–J) Comparisons of the numbers and areas of ducts among different genotypes. Shown are representative images from hematoxylin and eosin (HE)-stained tissue sections (Magnification: 100×) from mammary glands at the age of 6-week old, the average number of ducts per mm2 (H), ductal area to gland area ratio (ductal coverage) (I), and the average cross-section size per duct (10−3 mm2) (all ducts were measured for each mouse) (J). HE-stained tissue sections (G) were used for these analyses. Each bar is the average from three mice for each genotype. * and ** indicate P<0.05 and P<0.01 respectively. For panels H–I, one-way ANOVA was used for statistical analysis, while univariate analysis (sum of squares: type I) was used for panel J.
Figure 5.
Cell proliferation activity is increased by the knockout of Atbf1 in mammary epithelial cells from 6-week old mice.
(A–B) Tissue sections of mammary ducts (A) and terminal end buds (TEBs) (B) were subjected to IF staining against Ki67. DAPI staining was used to show nuclei (blue) and Ki67 was labeled by RITC (red). (C) Ratios of Ki67-positive cells to total epithelial cells in mammary glands with different Atbf1 genotypes were calculated. Each bar indicates the average result from 10 randomly selected areas. Three mice were used for each genotype, and the number of mammary epithelial cells counted for each bar was 1200–1600 for ducts and 2000–2800 for TEBs. *, P<0.05 and **, P<0.01 (one-way ANOVA).
Figure 6.
Atbf1-knockout-mediated cell proliferation primarily occurs in ER-positive cells.
(A, B) Detection of ER (red) and Ki67 (green) by double IF staining in mammary ducts (A) and TEBs (B). White arrows indicate cells that are positive for both ER and Ki67. (C, D) Knockout of Atbf1 increases cell proliferation in ER-positive cells (C) but not in ER-negative cells (D). (E) Knockout of Atbf1 also increases the number of ER-positive cells among proliferating (Ki67-positive) cells. For panels C–E, cells were counted from 10 randomly selected fields per mouse, and three mice were used for each genotype. The number of cells counted for each bar ranged from 1000 to 2000 for ducts and 2000 to 2500 for TEBs. *, P<0.05; **, P<0.01; n.s., P>0.05 (one-way ANOVA). (F) Real time RT-PCR was used to detect ER target gene expression (Areg, Igf-1, Ebag9 and c-Myc).
Figure 7.
Knockout of Atbf1 downregulates the expression of basal cell markers in 6-week mammary epithelial cells.
(A, B) Detection of CK5 and CK18 by IF staining in mouse mammary tissues. CK5 or CK18 was labeled with FITC (green), and DAPI staining shows nuclei. (C–D) Detection of mRNA levels for basal cell markers CK5, CK14 and CD44 (C) and luminal cell markers CK18, CK8 and CD24 (D) by real time RT-PCR in mammary tissues. Gapdh was used as the normalization control, and Atbf1 served as a positive control. Three mice were used for each genotype. *, P<0.05 and n.s., P>0.05 (one-way ANOVA).