Fig 2: BMPR1a regulated P-cadherin expression via p63 and Slug. (A) Immunohistochemistry staining for P-cadherin in control (n = 4 mice) and cKO (n = 3 mice) mammary glands at pregnancy day 14.5. Scale bar, 50 µm. (B) Western blotting for P-cadherin in mammary epithelial cells isolated from control and cKO mice at pregnancy day 14.5. ß-Tubulin was used as a loading control. Statistical analysis the expression of P-cadherin/ß-Tubulin. n = 3 mice. (C) qRT-PCR analysis of Cdh3 in FACS-sorted control and cKO myoepithelial cells at pregnancy day 14.5. n = 4 biological replicates. (D) P-cadherin (red) and K14 (green) double immunofluorescence staining in HC11 mammary epithelial cells treated with BMP4 (50 ng/mL) for 24 h. n = 3 biological replicates. Scale bar, 25 µm. (E) Western blotting for P-cadherin in HC11 mammary epithelial cells treated with BMP4 (50 ng/mL) for 24 h. ß-Tubulin was used as a loading control. Statistical analysis the expression of P-cadherin/ß-Tubulin. n = 3 biological replicates. (F) Scatter plot showing the correlation between BMPR1A and CDH3 expression in mammary glands from TCGA and GTEx data. Pearson’s coefficient test was performed to assess statistical significance. (G) Western blotting for P-cadherin in HC11 cells treated with p63 siRNA (sip63)/Slug siRNA (siSlug) and scramble RNA (NC) at 48 h. ß-Actin was used as a loading control. Statistical analysis the expression of P-cadherin/ß-Actin. n = 3 biological replicates. (H) Immunofluorescence for P-cadherin (green) in HC11 cells treated with p63 siRNA (sip63)/Slug siRNA (siSlug) and scramble RNA (NC) at 48 h. n = 3 biological replicates. Scale bar, 25 µm. (I) Western blotting for K14, P-cadherin, p63 and Slug in HC11 cells treated with scramble RNA, p63 siRNA, Slug siRNA, or p63 siRNA and Slug siRNA at 48 h. ß-Actin was used as a loading control. Statistical analysis the expression of K14/ß-Actin, P-cadherin/ß-Actin, p63/ß-Actin and Slug/ß-Actin. n = 3 biological replicates. Data were presented as means ± SD. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig 3: BMPR1a deficiency resulted in reduced myoepithelial lineages and compromised repopulating capacity of mammary stem cells. (A) FACS analysis of mammary gland epithelial cells from control and cKO mice at pregnancy day 14.5 using the Lin/CD24/CD29/CD61 surface marker set. n = 4 mice per group. (B) Quantification of Lin– CD24+CD29high (Myo), Lin– CD24+CD29low (Lu) and Lin– CD24+CD29lowCD61+ cell populations in control and cKO mice at pregnancy day 14.5. n = 4 biological replicates. (C) Immunofluorescence staining for K14 (red) and K8 (green) in control and cKO mammary glands at pregnancy day 14.5. n = 3 mice per group. Scale bar, 25 µm. (D) Western blotting for K14 in control and cKO mammary glands at pregnancy day 14.5. ß-Actin was used as a loading control. Statistical analysis the expression of K14/ß-Actin. n = 3 mice. (E) Immunofluorescence staining for K14 (red) and K8 (green) in HC11 mammary epithelial cells treated with BMP4 (50 ng/mL) for 24 h. n = 3 biological replicates. Scale bar, 25 µm. (F) Western blotting for K14 in HC11 cells treated with BMP4 (50 ng/mL) for 24 h. ß-Actin was used as a loading control. Statistical analysis the expression of K14/ß-Actin. n = 4 biological replicates. (G) Illustration of the mammary transplantation assays using K14-rtTA/teto-Cre/Bmpr1afl/fl and control mouse mammary stem cells and the induction strategy. (H) Whole-mount staining of mammary glands from the mammary transplantation assays and quantification of the proportion of ductal trees. n = 5 mice per group. Scale bar, 1 mm. (I) Immunohistochemistry staining for Integrin ß3 in control and cKO mammary glands at pregnancy day 14.5. Quantification of Integrin ß3+ cells in alveoli. n = 3 mice per group. Scale bar, 50 µm. Data were presented as means ± SD. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig 4: Working model of the role of BMPR1a in regulating mammary myoepithelial cells.

Fig 5: Loss of BMPR1a in myoepithelium resulted in an upregulation of Spp1. (A) qRT-PCR analysis of Fgf2, Hgf, Nrg1, and Spp1 in sorted mammary myoepithelial cells from control and cKO mice at pregnancy day 14.5. n = 4 biological replicates. (B) qRT-PCR analysis of Spp1 in sorted myoepithelial and luminal cells from wild-type mice at virginal stage and pregnancy day 14.5 (P14.5). n = 3 biological replicates. Data were presented as means ± SD. **p < 0.01, ***p < 0.001.