Beta1 integrin deletion from the basal compartment of the mammary epithelium affects stem cells

Abstract

The mammary gland epithelium comprises two major cell types: basal and luminal. Basal cells interact directly with the extracellular matrix (ECM) and express higher levels of the ECM receptors, integrins, than luminal cells. We show that deletion of beta1 integrin from basal cells abolishes the regenerative potential of the mammary epithelium and affects mammary gland development. The mutant epithelium was characterized by an abnormal ductal branching pattern and aberrant morphogenesis in pregnancy, although at the end of gestation, the secretory alveoli developed from beta1 integrin-positive progenitors. Lack of beta1 integrin altered the orientation of the basal-cell division axis and in mutant epithelium, in contrast to control tissue, the progeny of beta1 integrin-null basal cells, identified by a genetic marker, was found in the luminal compartment. These results reveal, for the first time, the essential role of the basal mammary epithelial cell-ECM interactions mediated by beta1 integrins in the maintenance of a functional stem cell population, mammary morphogenesis and segregation of the two major mammary cell lineages.

Figure 1. Abnormal ductal morphogenesis in the mammary epithelium presenting conditional deletion of the β1 integrin gene in basal epithelial cells

(a) Real time RT-PCR analysis of K18, K14 and Cre expression in basal and luminal cell populations isolated from 12-week-old virgin K5-Cre mouse mammary gland. Data are presented as means ± S.E.M. obtained in two independent experiments. (b) and (c) X-gal whole-mount staining. (b) Mammary rudiments from 3-week-old control (K5Cre;itgb1^F/+) and mutant (K5Cre;itgb1^F/F) mice; (c) outgrowths developed by control (K5Cre;itgb1F/+) and mutant epithelium in virgin recipient mouse. (d) and (e) Immunofluorescence staining of the sections through control and K5Cre;itgb1^F/F outgrowths with anti-K5, anti-β1 integrin (d), anti-cleaved caspase 3 and α-SM-actin (α-SMa) (e) antibodies. Arrows in (d) indicate the basal cell layer. Arrows in (e) indicate cleaved caspase 3- and α-SM-actin-positive basal cells. DAPI served to stain nuclei. (f) Histograms showing ratio between K5-positive and K5-negative mammary epithelial cells in control and mutant outgrowths developed in virgin host. Data are presented as means ± S.E.M. of at least five mammary ducts from two animals in each case. Scale bars, 3 mm in (b) and (c), upper panel, 0.7 mm in (c), lower panel, and 55 μm in (d) and (e).

Figure 2. Lack of functional stem cells in K5Cre;itgb1^F/F epithelium

(a) Whole-mount X-gal staining of the secondary outgrowths produced by control K5Cre;itgb1^F/+ and mutant K5Cre;itgb1^F/F epithelium in the cleared fat pads of virgin recipient mice. The outgrowths were analysed 10 weeks after transplantation. Arrow (upper panel) indicates small pieces of transplanted mutant epithelium that did not develop any secondary outgrowth. Lower panel shows the most developed mutant outgrowth and corresponding control. Scale bar, 3 mm. (b) Flow cytometry analysis of mammary epithelial cells isolated from outgrowths developed by control (upper panels) and mutant (lower panels) epithelium in 12-week-old virgin recipient mice. Cells were stained for CD24 and β1 integrin (left) or CD24 and α6 integrin expression (right). Only CD45⁻CD31⁻ cells were included in the analysis. Red ellipses show CD24-positive/β1-high and CD24-positive/α6-high cell populations. The percentages of β1- and α6-negative (left to vertical reference line) and positive (right to vertical reference line) cells were calculated for the CD24- positive population (above horizontal reference line) comprising mammary epithelial cells.

Figure 3. Perturbation of lobuloalveolar development in K5Cre;itgb1^F/F mammary epithelium

(a) Whole-mount X-gal staining of the K5Cre;itgb1^F/+ and K5Cre;itgb1^F/F outgrowths developed in 7.5-day-pregnant host. (b) H&E staining of the sections through control and K5Cre;itgb1^F/F outgrowths from 14.5-day-pregnant host. Arrows and arrowheads indicate alveoli and TEB-like structures, respectively. (c) Quantitative RT-PCR analysis of β-casein and WAP expression in mammary glands from 13.5-daypregnant control and K5Cre;itgb1^F/F mice. The values were normalised to K18 expression. Data are presented as means ± S.E.M. obtained in two independent experiments. (d-f) Double immunofluorescence staining of the sections through control and K5Cre;itgb1^F/F outgrowths developed in 14.5- (d and e) and 7.5-day- day pregnant host (f) with anti-β1 integrin (d, e and f), anti-K5 (d), anti-α-SM-actin (e), anti-K8 (f) antibodies. Arrows in (d and e) indicate basal cell layer, asterisks show position of luminal layer. Scale bars, 0.77 mm in (a); 160 μm in (b), and 55 μm in (d–f).

Figure 4. Altered orientation of the basal cell division axis in K5Cre;itgb1^F/F mammary epithelium

(a) Whole-mount X-gal staining of the K5Cre;itgb1^F/+ and K5Cre;itgb1^F/F outgrowths developed in 7.5-dpc host. Arrows show LacZ-negative (pink) cells. (b) Dividing basal cells in the ducts formed by control and K5Cre;itgb1^F/F epithelium. Double immunofluorescence staining with anti-K5 and anti-β-tubulin antibodies. Basement membrane position is marked by discontinuous lines, double-headed arrows indicate division plane. (c) Position of ductal basal cell division plane in the mammary outgrowths developed in host mice at 7.5 dpc. The numbers in green and red correspond to the numbers of cells dividing parallel and perpendicular to the basement membrane, respectively. The thickness of colored bars is proportional to the cell number. Cell counts obtained for each of the four mice used for the analysis are presented in Supplementary Information, Table 1. Scale bars, 100 μm (a), and 75 μm (b).

Figure 5. K5-negative progenitors give rise to functional alveoli late in pregnancy

(a) H&E staining of the section through outgrowths developed by control and K5Cre;itgb1^F/F epithelium in a 18.5-day-pregnant host. Arrows indicate well-developed alveoli in mutant epithelium; arrowheads and asterisks mark TEB-like structures and collapsed alveoli, respectively, persisting in K5Cre;itgb1^F/F outgrowths. (b) X-gal staining of the sections through K5Cre;itgb1^F/F outgrowths developed in 16.5- and 18.5-day-pregnant host. Arrows show newly formed alveoli consisting essentially of LacZ-negative cells. (c) Double immunofluorescence staining of the sections through control and K5Cre;itgb1^F/F outgrowths developed in 18.5-daypregnant host with antibodies against β1 integrin and K5. (d) Whole-mount X-gal staining of secondary outgrowths resulting from re-transplantation of pieces of control (K5Cre;itgb1^F/+) and mutant (K5Cre;itgb1^F/F) epithelium developed in 15.5 daypregnant host mouse. In six out of seven grafted mice, control epithelium produced secondary outgrowths occupying the entire mammary fat pad, one outgrowth occupied approximately 30% of the fat pad. Mutant epithelium did not develop in five out of seven fat pads and produced two small outgrowths occupying approximately 10–15% of the fat pad. Scale bars, 0.3 mm in (a), 100 μm in (b), 75 μm in (c) and 3 mm in (d).