Lineage-Biased Stem Cells Maintain Estrogen-Receptor-Positive and -Negative Mouse Mammary Luminal Lineages

Abstract

Delineating the mammary differentiation hierarchy is important for the study of mammary gland development and tumorigenesis. Mammary luminal cells are considered a major origin of human breast cancers. However, how estrogen-receptor-positive (ER⁺) and ER^- luminal cells are developed and maintained remains poorly understood. The prevailing model suggests that a common stem/progenitor cell generates both cell types. Through genetic lineage tracing in mice, we find that SOX9-expressing cells specifically contribute to the development and maintenance of ER^- luminal cells and, to a lesser degree, basal cells. In parallel, PROM1-expressing cells give rise only to ER⁺ luminal cells. Both SOX9⁺ and PROM1⁺ cells specifically sustain their respective lineages even after pregnancy-caused tissue remodeling or serial transplantation, demonstrating characteristic properties of long-term repopulating stem cells. Thus, our data reveal that mouse mammary ER⁺ and ER^- luminal cells are two independent lineages that are maintained by distinct stem cells, providing a revised mammary epithelial cell hierarchy.

Keywords: breast cancer; cancer cell-of-origin; estrogen receptor negative cancer; estrogen receptor positive cancer; lineage tracing; mammary differentiation; mammary gland development; mammary gland regeneration; mammary stem cells; stem cell hierarchy.

Figure 1. SOX9⁺ cells are long-term repopulating and maintain the ER⁻ luminal and basal lineages in the postnatal mammary gland

(A) The mouse models (upper) and the experimental schedule (lower) for lineage-tracing experiments shown in Figure 1B to 1F. (B) Representative tdTomato whole-mount images of Sox9-CreER^T2; R26R-tdTomato mammary glands at the indicated time points after tamoxifen treatment. Mice without tamoxifen treatment (No TAM) at 24 weeks of age were used as the negative control (inset). Minimum 3 animals were examined for each time point. (C) Percentages of tdTomato⁺ cells in the basal and luminal populations at the indicated time points, as determined by flow cytometry (mean ± SEM, n = 3–8). (D) K8 and K14 immunostaining of mammary gland sections for characterizing the tdTomato⁺ cells after 20 weeks of lineage tracing. Representative luminal only and bi-lineage tdTomato⁺ ducts are shown. The arrow head and arrows point to basal and luminal cells, respectively. Total 57 tdTomato⁺ cell clusters in 3 mice were counted. (E) Percentages of tdTomato⁺ cells in the Sca1⁺ and Sca1⁻ luminal populations at the indicated time points, as determined by flow cytometry (mean ± SEM, n = 2–3). (F) ER and PR immunostaining of mammary gland sections 2 weeks after tamoxifen treatment. Magnification of the selected areas is shown on the right of each panel. 98% of tdTomato⁺ cells were ER⁻ and PR⁻. (G) Experimental schedule (left) for adult Sox9-CreER^T2 lineage-tracing experiments as shown in Figure 1G to 1I, and a representative tdTomato whole-mount image (right) of Sox9-CreER^T2; R26R-tdTomato mammary glands 20 weeks after the tamoxifen treatment. (H) Percentages of tdTomato⁺ cells in the basal, Sca1⁺ and Sca1⁻ luminal populations at indicated time points after the tamoxifen treatment, as determined by flow cytometry (mean ± SEM, n = 3–6). (I) ER immunostaining of mammary glands after 20 weeks of lineage tracing of adult SOX9⁺ cells. 99.5% tdTomato⁺ cells were ER⁻. *P < 0.05, **P < 0.01, ***P<0.001, and ****P < 0.0001. See also Figure S1.

Figure 2. SOX9⁺ cells contribute to alveologenesis and persist after repeated pregnancy

(A) The experimental schedule for lineage-tracing studies shown in Figure 2. (B) A representative tdTomato whole-mount image of Sox9-CreER^T2; R26R-tdTomato mammary glands on day 17.5 of pregnancy (Preg17.5d). (C) Percentages of tdTomato⁺ cells in the basal and luminal populations of Preg17.5d mice, as determined by flow cytometry (mean ± SEM, n = 3, paired t-test). (D) K8 and K14 immunostaining of Preg17.5d mammary gland sections for characterizing the tdTomato⁺ cells. Examples of alveoli with either tdTomato-labeled luminal or basal cells were shown. The graph on the right shows percentage of alveoli containing the indicated types of tdTomato⁺ cells. (E) ER immunostaining of Preg17.5d mammary gland sections. 99% tdTomato⁺ cells were ER⁻. (F) A representative tdTomato whole-mount image of Sox9-CreER^T2; R26R-tdTomato mammary glands at least 3 weeks after weaning of the second litter. (G) Percentage of tdTomato⁺ cells in the basal, Sca1⁺ and Sca1⁻ luminal populations 3 weeks after weaning of the second litter (mean ± SEM, n = 6). (H) ER immunostaining of mammary glands after the 2^nd pregnancy. 99.5% of tdTomato⁺ cells were ER⁻. *P < 0.05 and **P < 0.01. See also Figure S2.

Figure 3. PROM1⁺ cells are long-term populating and specifically maintain the ER⁺ lineage

(A) The mouse models (upper) and the experimental schedule (lower) for PROM1⁺ cell lineage-tracing experiments as shown in Figure 3A–3D. (B) Representative tdTomato whole-mount images of Prom1-CreER^T2; R26R-tdTomato mammary glands at the indicated time points after tamoxifen treatment. Mice without tamoxifen treatment (No TAM) at 24 weeks of age were used as the negative control. Minimum 3 animals were examined for each time point. (C) Representative flow cytometric profiles of the basal, Sca1⁺ and Sca1⁻ luminal populations showing percentages of tdTomato⁺ cells after 20 weeks of lineage tracing. Percentages of tdTomato⁺ cells in specific cell populations are quantified on the right (mean ± SEM, n = 3–6, paired t-test). (D) ER and PR immunostaining of mammary gland sections after 20 weeks of lineage tracing. 97% and 100% of tdTomato⁺ cells are ER⁺ and PR⁺, respectively. (E) The experimental schedule for lineage-tracing studies of Prom1-CreER^T2; R26R-tdTomato mice during pregnancy (left), and ER immunostaining of mammary gland sections at Preg17.5d (right). All tdTomato⁺ cells were ER⁺. (F) A representative tdTomato whole-mount image of Prom1-CreER^T2; R26R-tdTomato mammary glands 3 weeks after 2 rounds of pregnancy and lactation. (G) Percentages of tdTomato⁺ cells in the indicated mammary cell types 3 weeks after 2 rounds of pregnancy and lactation, as determined by flow cytometry (mean ± SEM, n = 6, paired t-test). (H) Percentages of tdTomato⁺ cells in the indicated mammary cell types at the indicated time points of adult PROM1⁺ cell lineage tracing (n = 2–3). (I) ER immunostaining of mammary gland sections after 8 weeks of adult PROM1⁺ cell lineage tracing. 96% tdTomato⁺ cells were ER⁺. ***P<0.001. See also Figure S3.

Figure 4. ER⁻ and ER⁺ luminal lineages are regenerated by different unipotent stem cells during serial mammary gland transplantation

(A) A schematic diagram of the serial transplantation experiments. Sox9-CreER^T2; R26R-tdTomato or Prom1-CreER^T2; R26R-tdTomato mice were treated with tamoxifen at P28. Small mammary ductal fragments were isolated ≥8 weeks later and used for cleared mammary fat pad transplantation. The fragments of primary outgrowths were then used for secondary transplantation. (B) Representative tdTomato whole-mount images of the primary and secondary outgrowths regenerated by Sox9-CreER^T2; R26R-tdTomato mammary ductal fragments. (C) Representative flow cytometric profiles and quantification of percentages of tdTomato⁺ cells in the indicated cell types of Sox9-CreER^T2; R26R-tdTomato primary outgrowths (mean ± SEM, n = 4). (D) ER and PR immunostaining of the secondary outgrowths from Sox9-CreER^T2; R26R-tdTomato ductal fragments. 97% and 99% of tdTomato⁺ cells were ER⁻ and PR⁻, respectively. (E) Representative tdTomato whole-mount images of the primary and secondary outgrowths derived from Prom1-CreER^T2; R26R-tdTomato mammary ductal fragments. (F) Representative flow cytometric profiles and quantification of frequency of tdTomato⁺ cells in the indicated cell types of Prom1-CreER^T2; R26R-tdTomato primary outgrowths (mean ± SEM, n = 7). (G) ER and PR immunostaining of the secondary outgrowths from Prom1-CreER^T2; R26R-tdTomato ductal fragments. 97% and 98% of tdTomato⁺ cells were ER⁺ and PR⁺, respectively. **P < 0.01 and ****P < 0.0001. See also Figure S4.