Figure 1.
Luminal and basal/myoepithelial layers in the bovine mammary gland can be distinguished by immunofluorescence analysis.
A: H&E staining of sections from a heifer's mammary gland reveals ductal structures penetrating the fibrous stroma. B–F: Immunofluorescence detection of mammary lineage markers. Inset: 2× magnification. L = lumen. Bar = 50 µm.
Figure 2.
Four populations of epithelial cells with distinct CD24 and CD49f expression were identified in the bovine mammary gland.
Lin− bMECs from heifer mammary gland were sorted according to CD24 and CD49f expression. Two main populations: CD24neg-medCD49fpos and CD24med-highCD49fneg, encircled by dashed green lines, emerged in the density plot. Putative populations enriched with stem cells (puStm, 5.8±1.3%) and their progenitors (puPgt, 8.0±1.4%), as well as their complementary Basal (11.7±2.9%) and luminal (Lum, 12.2±2.6%) populations (encircled by solid red lines) were collected. Inset: gating of living cells (framed in red) according to PI staining. Percentage of each population was calculated out of the total living cells detected. FSC – forward scatter.
Figure 3.
Delineation of bMEC populations by expression analysis.
Expression levels of selected genes, relative to ungated Lin− cells, were determined in the four sorted bMEC populations by real-time PCR. A: Differential mRNA expression of basal (CD49f, CK14, p63 and CK6) and luminal (CK18) markers infers the location of the sorted populations within the mammary tissue. B: Differential expression of genes implicated as stem and progenitor cell markers in several adult tissues. C: Differential expression of genes associated with luminal lineage. D: Differential expression of genes along the Notch pathway. Columns represent mean±SEM of data collected from four individual heifers and different letters above the columns indicate statistically significant (P<0.05) differences in the comparison of each value to its three counterpart values.
Figure 4.
In-situ localization of proteins with distinct expression among the bMEC populations.
All analyses depict immunofluorescence detection, except for ALDH1 which was detected by DAB reaction, generating a brown signal with hematoxylin counterstaining of the nuclei. ALDH1: red arrows mark positively stained cells in the stromal area. Notch1: red arrows mark positively stained cells in the basal layer. Bar = 50 µm.
Figure 5.
Multipotency, high propagation potential and clone-formation capability characterize the puStm population.
A: Percentage of colony-forming cells out of the total sorted cell population. B: Immunofluorescence staining of adherent clones, demonstrating two types: luminal clones expressing CK18 (left) and basal clones expressing CK14 (right). C: Composition of clone types formed by each of the sorted populations. Numbers of defined clones are relative to their total number. D: Differences in propagation were observed among the sorted populations during the first 7 days in culture. Columns represent mean±SEM of three analyses. Different letters above the columns indicate statistically significant (P<0.05) differences. Bar = 50 µm.
Figure 6.
Cultured bMEC populations do not differ in their CD49f/CD24 expression, but maintain their distinct parental characteristics.
A: Schematic representation of the experimental procedure. B: FACS histograms depicting the levels of CD24 and CD49f in freshly isolated bMECs compared with their cultured counterparts. C. FACS dot-plots depicting the subpopulations sorted from the cultured cells. D: Immunofluorescence staining of the lineage markers CK14 and CK18 in organized and non-organized colonies. E: Regardless their different cell-surface marker expression, organized colonies were significantly more frequent in sorted cultured cells originated from the puStm and Basal populations. Bar = 50 µm.
Figure 7.
Mammospheres are formed by freshly dissociated bMECs, whereas sorting procedures induce non-spherical floating colonies (NSFCs).
A: Representative mammosphere, formed by freshly dissociated Lin− bMECs. Nuclei are stained with DAPI. B: Supplementation of conditioned mammary medium enhances mammosphere formation. *,**Significantly different at P≤0.05 and P≤0.01, respectively. C: The mammosphere is comprised of cells expressing CK14 and CK18. D: Antibody labeling and the sorting process prevent mammosphere formation. E: Representative NSFC formed by sorted bMECs. F: NSFC is comprised mainly of live cells. Trypan blue-stained dead cell (blue) is marked by an arrow. G: The NFSC is comprised of cells expressing CK14 and CK18. Columns represent average±SEM of three wells analyzed for each group. Bar = 50 µm.
Figure 8.
NSFC development and characteristics depend on its origin.
A: Representative demonstration of limited development of a cultured NSFC from Lum cells compared to NSFCs from the other populations. B: Serial dissociation and culture of NSFCs over three generations demonstrates limited self-renewal capacity and a more severe effect of sorting, compared with antibody labeling, on their development. C: puStm and puPgt cultures generate higher numbers of NSFCs compared with Basal and Lum cultures. Columns represent mean±SEM of three analyses of least 26 floating colonies for each population. Different letters above the columns indicate statistically significant (P<0.05) differences. Bar = 50 µm.
Figure 9.
Incorporating ALDH activity into the CD24/CD49f-based analysis reveals a small ALDHbr population within the puStm fraction that is potentially enriched with stem cells.
A: FACS analysis of Lin− bMECs and gating of ALDH-positive (ALDHbr) cells according to the effect of the ALDH inhibitor DEAB. B: Demonstration of ALDHbr (red) and ALDHneg (green) distribution among the populations sorted according to CD49f and CD24 expression. SSC - side scatter.
Figure 10.
Proposed bovine mammary epithelial cell hierarchy.
Table 1.
List of antibodies used in the study.