Roles of Fas and Fas ligand during mammary gland remodeling (original) (raw)
Fas/FasL mRNA expression in the mouse mammary gland. Terminal differentiation of the mammary gland normally takes place only during pregnancy and lactation. Gestational and lactogenic hormones regulate this pathway of differentiation. Changes in the hormonal environment at the end of lactation trigger the apoptotic signal required for “breast remodeling”. We tested the hypothesis that the Fas/FasL system could mediate apoptotic signals during tissue remodeling by studying the expression of Fas and FasL in mice mammary tissue. Studying mice, we analyzed Fas and FasL expression at the mRNA and protein level. RT-PCR was performed on samples (n = 3/group) obtained from nonpregnant virgin mice during pregnancy days 15 and 19, after 10 days of lactation, and days 1, 2, 4, and 12 after weaning. The intensity of the signal was analyzed by densitometry as described in Methods and plotted as a bar in Figure 1. Fas mRNA was clearly detectable in all tissues studied (Figure 1). In contrast, FasL mRNA was absent in nonpregnant, virginal mice (Figure 1, lane 1), began to appear by day 15, and was greatest on day 19 of pregnancy (Figure 1, lanes 2 and 3). During lactation and the first day of weaning, FasL mRNA was also present in high levels (Figure 1, lanes 4 and 5), but decreased dramatically by postlactational days 4 and 12 (Figure 1, lanes 6–8).
Expression of Fas/FasL mRNA in mouse mammary tissue. RT-PCR for Fas/FasL was performed with total RNA extracted from BALB/c mice during pregnancy, lactation, and involution. FasL mRNA was expressed during pregnancy, lactation, and early involution, but not in breast tissue from virgin mice. Fas mRNA was present in all the tissues evaluated. The intensity of the signal was analyzed using a digital imaging analysis system (AlphaEase). Bars represent the average ± SD from three independent experiments. The figure of the gel is a representative experiment of the three experiments. Lane 1, virgin; lane 2, day 15 of pregnancy; lane 3, day 19 of pregnancy; lane 4, day 10 of lactation; lane 5, day 1 after weaning (AW); lane 6, day 2 AW; lane 7, day 4 AW; lane 8, day 12 AW.
Western blot analysis for Fas, FasL, and caspase 3 expression in the mouse mammary gland. To compare the expression of Fas and FasL mRNA with protein expression we performed Western blots on protein preparations from the same samples we employed for mRNA analysis. The intensity of the staining was analyzed by densitometry as described above. As shown in Figure 2, Fas and FasL protein expression differed strikingly from their patterns of mRNA expression. Mammary glands from virgin mice express high protein levels of Fas, but not FasL, protein (Figure 2, lane 1). However, during pregnancy and lactation, Fas protein expression was suppressed (Figure 2, lanes 2–4), while FasL was expressed in high amounts compared with the nonpregnant mouse mammary gland (Figure 2, lanes 2–4).
Expression of Fas/FasL protein in mouse mammary tissue. Western blot analysis for Fas and FasL was performed with protein samples obtained from the same samples described in Figure 1. Fas expression is found in breasts from virgin mice but is not expressed during pregnancy and lactation. High levels of Fas protein appear 1 day after weaning. FasL is expressed during pregnancy, lactation, and involution, but not in the virgin mouse. The intensity of the signal was analyzed using a digital imaging analysis system. Bars represent the average ± SD from three independent experiments. The figure of the immunoblot is a representative experiment of the three experiments. Lane 1, virgin; lane 2, day 15 of pregnancy; lane 3, day 19 of pregnancy; lane 4, day 10 of lactation; lane 5, day 1 AW; lane 6, day 2 AW; lane 7, day 4 AW; lane 8, day 12 AW.
Removal of the litters, which constitute the primary factor for mammary gland involution, had a profound effect on Fas expression; 24 hours after the removal of the litters, Fas protein level increased dramatically (Figure 2, lane 5). Quantification of the signal indicated a statistically significant increase (P < 0.001) of Fas protein levels on days 1, 2, 4, and 8 of involution compared with the protein levels detected during pregnancy (Figure 2, lanes 5, 6, 7, and 8, respectively). FasL protein levels remained high during the first two days of involution and were significantly reduced from days 4 to 12 after weaning (P < 0.05) (Figure 2, lanes 5, 6, 7, and 8, respectively). Similarly, the active form of caspase 3 was detected by Western blot analysis during mammary gland remodeling. Thus, high concentration of the active form of caspase 3 was mainly detected during the first 4 days of involution (Figure 3, lanes 3–6), decreasing afterward (Figure 3, lanes 7 and 8). On the other hand, the active form of caspase 3 was not expressed in the mammary tissue from nonpregnant/virgin mice (Figure 3, lane 1).
Expression of caspase 3 in mouse mammary tissue. Western blot analysis for caspase 3 was performed with protein samples obtained as described in Figure 1. Increase in the expression of the active form of caspase 3 was detected 24 hours after weaning; expression remained high during days 2 and 4 of involution, decreasing afterward. Lane 1, virgin; lane 2, lactation 10 days; lane 3, day 1 AW; lane 4, day 2 AW; lane 5, day 3 AW; lane 6, day 4 AW; lane 7, day 8 AW; lane 8, day 12 AW.
Localization of FasL and Fas expression in the normal mouse mammary gland. To determine which cells expressed Fas and FasL, we performed immunohistochemistry on paraffin sections of the mammary gland from normal mice during pregnancy, lactation, and weaning. Sections were stained with a polyclonal Ab for FasL (clone N-20; Santa Cruz Biotechnology Inc.) and for Fas using the mAb B-10 (Santa Cruz Biotechnology Inc.). Mammary gland tissues were collected from virgin mice and during different days of pregnancy (as described in Methods), lactation, and involution. Mammary gland samples from virgin mice were immune negative for FasL (Figure 4a), but positive for Fas (Figure 5a). However, at days 15 and 19 of pregnancy abundant FasL-positive immunoreactive cells were found in the glandular epithelium (Figure 4b). The number and intensity of positive immunoreactivity for FasL increased significantly as pregnancy progressed (from day 5 to day 19, data not shown). Immunoreactivity for Fas was almost absent, with the exception of a few scattered positive stromal cells with the cytology of macrophages and leukocytes (Figure 5b, and data not shown). After delivery and during lactation, few Fas-positive glandular epithelial cells were found, while FasL immunoreactivity was present in almost all of the glandular epithelium, including ducts (data not shown).
Localization of FasL expression in mouse mammary tissue during pregnancy, lactation, and involution. Paraffin sections of normal breast tissue were stained with anti-mouse FasL Ab. (a) Virgin mouse breast, showing immune negative for FasL. (b) Day 15 of pregnancy, abundant positive FasL immune reactivity was shown in the glands. (c) Day 1 of involution, FasL immunoreactivity is present in glandular epithelial cells surrounding milk accumulation. (d) Day 2 of involution, the number of FasL-expressing cells increased. (e) Day 3 of involution. (f) Day 5 of involution, characterized by presence of ducts and epithelial chordae surrounded by adipocytes with localization of FasL in a few cells. ×400.
Localization of Fas expression in mouse mammary tissue during pregnancy, lactation, and involution. Immune staining of paraffin sections from mouse breast tissue is shown. (a) Virgin mouse breast, showing Fas-positive immune reactivity. (b) Day 15 of pregnancy, Fas-positive immune reactivity was absent except for some scattered positive stromal cells. (c) Day 1 of involution, strong Fas-positive immune reactivity was found in part of the glandular epithelium. (d) Day 2 of involution, the number of Fas-expressing cells is significantly increased. (e) Day 3 of involution. (f) Day 5 of involution, showing mainly ducts and epithelial chordae surrounded by adipocytes with localization of Fas in a few cells. ×400.
Localization of FasL and Fas expression during mammary gland involution. Postlactational involution of the mammary gland during weaning is characterized by cell death of portions of the mammary epithelium. To determine how Fas and FasL expression is related to mammary gland involution we stained paraffin sections from mouse mammary glands obtained consecutive days after weaning. The number of glandular epithelial cells expressing FasL and Fas increased dramatically between day 1 and 2 after weaning (Figure 4, c and d; Figure 5, c and d). By day 4, the tissue was characterized by the presence of mainly ducts and clusters of epithelial chordae surrounded by adipocytes. Although FasL was still present, it was expressed in fewer cells than on previous days (Figure 4f).
Presence of apoptotic cells in the mammary epithelium of normal mice, but not in the Fas-deficient lpr mice. To determine if the expression of Fas and FasL is related to the apoptosis of the mammary epithelium characteristic during involution, we performed histological analysis by using the TUNEL assay on paraffin sections derived from mammary glands at days 1, 2, 4, and 12 of involution in normal BALB/c mice. Similarly, we compared the presence of apoptotic cells in the involuting mammary gland of MRL mice, the Fas-deficient MRL/lpr mice, as well as of C3H and C3H/gld/gld (FasL-deficient) mice. When we analyzed DNA fragmentation in mammary glands from MRL/lpr or C3H/gld/gld mice, no apoptotic cells were detected during the first 3 days of involution (Figure 6, c and e). However, apoptotic cells were present at day 4 after weaning in the MRL/lpr mice (Figure 6d) and in the C3H/gld/gld mice (Figure 6f). In contrast, DNA fragmentation (shown as blue nuclei) was present at day 1 of involution in MRL (Figure 6a) and C3H mice and increased to a substantial number at day 2 of involution, reaching a relative maximum at day 4 after weaning (Figure 6b).
Localization of apoptosis in mouse mammary tissue during involution. Apoptosis in the involuting mammary gland was monitored by TUNEL on 5-μm paraffin sections derived from mammary glands of the MRL and C3H mouse, the Fas-deficient MRL/lpr mouse, and the FasL-deficient C3H/gld mouse. (a) Wild-type MRL mouse, day 1 of involution, showing numerous apoptotic cells (positive nuclei are detected as blue spots) in glandular structure. (b) Wild-type MRL mouse, day 4 of involution; the number of apoptotic cells is significantly increased. (c) MRL/lpr mouse, day 1 of involution, showing no evidence of apoptosis on intact glandular structure. (d) MRL/lpr mouse, day 4 of involution, showing increased number of apoptotic cells. ×400. (e) C3H/gld mouse, day 1 of involution, showing no evidence of apoptosis on intact glandular structure. ×200. (f) C3H/gld mouse, day 4 of involution, showing apoptotic cells. ×200.
In vitro hormonal regulation of Fas and FasL expression in mammary epithelial cells. To establish an in vitro model to study the role of the Fas/FasL system in mammary epithelial cell–mediated apoptosis, we performed studies using either the normal mouse epithelial cell line HC-11 or the normal human epithelial MCF-10A cells (data not shown).
Effect of hormone withdrawal on Fas and FasL expression by normal mouse HC-11 mammary epithelial cells. To study the hormonal involvement in Fas/FasL expression during involution of the mouse mammary gland, we attempted to reproduce in vitro conditions resembling normal involution after weaning. The mouse mammary glandular epithelial cell line HC-11 was cultured in DMEM-F12 with 10% FCS and containing dexamethasone (1 μM), insulin (5 μg/ml), and prolactin (5 μg/ml) to induce lactogenic differentiation as described previously (30). We used a JAM or a MTT assay to evaluate the effect of hormone and serum withdrawal on Fas and FasL expression. Both methods gave similar results. As shown in Figure 7, HC-11 cells grown in normal conditions express only very low levels of Fas protein (Figure 7a, lane 1) and do not express FasL protein (Figure 7b, lane 1). By 1 hour after removal of the serum and hormones from the media (SM), there was a fourfold increase in Fas protein levels, and they remained high for up to 24 hours (Figure 7a, lanes 2–5). Interestingly, FasL protein expression increased only after 2 hours of incubation in SM and reached its highest level measured after 24 hours (Figure 7b, lanes 3–5).
Expression of Fas/FasL and caspase 3 in mouse mammary glandular epithelial cell line HC-11. Western blot analysis for Fas (a) and FasL (b) was performed with protein extracted from in vitro culture of HC-11 cells. Note the increase of Fas protein expression after 1 hour in hormone- and serum-deprived media (SM). FasL protein was expressed after 2 hours in SM and reached the highest level after 24 hours. (c) Western blotting for caspase 3 was performed with the same protein as above. Presence of SM converted the inactive form of caspase 3 to an active form corresponding to Fas and FasL expression. Lane 1, culture in normal media; lane 2, 1 hour in SM; lane 3, 2 hours in SM; lane 4, 6 hours in SM; lane 5, 24 hours in SM.
Effect of SM treatment on caspase 3 expression. Fas/FasL–induced apoptosis is mediated by the activation of a proteolytic system involving proteases called caspases. The activation of caspase 3 is an important component of this cascade. To further confirm the involvement of the Fas/FasL system in the induction of apoptosis of mammary epithelial cells, we studied caspase 3 expression in HC-11 cells after removal of serum and hormones. As shown in Figure 7c, HC-11 cells maintained in normal conditions express only the inactive form of caspase 3 (Figure 7c, lane 1); incubation in SM induced the activation of caspase 3 in HC-11 cells after 1 hour (corresponding to Fas and FasL expression). As apoptosis progressed, the amount of the inactive form of caspase 3 decreased (Figure 7c, lanes 2–5).
Correlation between Fas/FasL expression and apoptosis in mammary epithelial cells. To compare the in vitro effect of hormonal deprivation on mammary epithelial cells’ Fas and FasL expression and activation of caspase 3 with apoptosis, we performed annexin V and JAM assays. Cells were treated with SM as described above followed by incubation with FITC-labeled annexin V. Cells were then observed under a fluorescent microscope using 4,6-diamidino-2-phenylindole (DIP) as marker for the presence of cells in the culture. As shown in Figure 8, numerous annexin V–positive cells were present after 24 hours (Figure 8d) and 48 hours (Figure 8f) in SM, but not in the control group (Figure 8b). To quantify the number of apoptotic cells that were the result of culture in SM, we used the JAM assay. We found 40% (± 12.9) and 43% (± 6) of apoptotic cells after 24 hours and 48 hours, respectively, in SM (Figure 9). Furthermore, to demonstrate the role of the Fas receptor in the regulation of HC-11 SM–induced apoptosis, cells were treated with anti-FasL (NOK-1 mAb) at the beginning of the culture. By blocking Fas/FasL interaction the apoptotic effect of hormonal removal in HC-11 cells was decreased (Figure 9). The control Ab, IgG1 isotype control immunoglobulin (clone 107.3; PharMingen) had no effect.
In situ detection of apoptotic cells in in vitro culture of mammary glandular epithelial cells. MCF-10A and HC-11 cells were treated with SM for 24 and 48 hours. Apoptotic cells were detected by annexin V (b, d, f). The figures are a representative experiment using HC-11 cells. (b) Annexin V staining of cells in complete media. (d) After 24 hours in SM. (f) After 48 hours in SM. (a, c, e) DAPI staining was used to prove the presence of cells in b, d, and f, respectively. ×400.
Fas dependence of SM-induced apoptosis. Mammary glandular epithelial cells MCF-10A (human) and HC-11 (mouse) were labeled with [3H]-thymidine for 24 hours and treated with SM for 24 hours and 48 hours in the presence or absence of anti-FasL (NOK-1 mAb, 100 ng/ml). As control we used the purified mouse IgG1 isotype control from PharMingen. The percentage of apoptosis was calculated as described in the text. Data represent the mean ± SD of three independent experiments. A_P_ < 0.001, B_P_ < 0.05.
Fas-induced apoptosis. Further confirmation that Fas expressed by mammary cells is active was demonstrated by in vitro experiments using SM media and the agonist anti-Fas mAb. Mammary epithelial cells (MCF-10A and HC-11) were treated with increasing concentrations of anti-Fas mAb for 24 hours, and the Fas-mediated apoptosis was determined by either JAM or MTT assay. Upregulation of Fas expression after treatment with SM media resulted in increased responsiveness toward Fas-mediated apoptosis (Figure 10).
Effects of the anti-Fas mAb on HC-11 and MCF-10A cells. Normal mammary glandular epithelial cells MCF-10A and HC-11 were cultured with different concentrations of anti-Fas IgG mAb for 24 hours (ranging from 5–500 ng/ml). MTT assay was performed to evaluate cell viability. As control, cells were treated with a mouse IgG1 mAb at concentrations similar to the anti-Fas Ab. A_P_ < 0.05. Results shown are representative of at least three independent experiments done with HC-11 cells.
In conclusion, culture of either HC-11 or MCF-10A cells in SM induced the rapid expression of Fas and FasL, which is followed by activation of caspase 3 and the induction of apoptosis in more than 40% of the cells. Furthermore, addition of DEVD-CHO (caspase inhibitor) protected mammary epithelial cells to the apoptotic effect of Fas Ab or SM (data not shown). DEVD-CHO is a pseudosubstrate that mimics the caspase-3 cleavage site and thus inhibits its protease activity.