(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Rac1 controls cell turnover and reversibility of the involution process in postpartum mammary glands [1]

['Aleksandr Mironov', 'Faculty Of Biology Medicine', 'Health', 'University Of Manchester', 'Manchester', 'United Kingdom', 'Matthew Fisher', 'The Bateson Centre', 'Department Of Oncology', 'Metabolism']

Date: 2023-01

Cell turnover in adult tissues is essential for maintaining tissue homeostasis over a life span and for inducing the morphological changes associated with the reproductive cycle. However, the underlying mechanisms that coordinate the balance of cell death and proliferation remain unsolved. Using the mammary gland, we have discovered that Rac1 acts as a nexus to control cell turnover. Postlactational tissue regression is characterised by the death of milk secreting alveoli, but the process is reversible within the first 48 h if feeding recommences. In mice lacking epithelial Rac1, alveolar regression was delayed. This defect did not result from failed cell death but rather increased cell turnover. Fitter progenitor cells inappropriately divided, regenerating the alveoli, but cell death also concomitantly accelerated. We discovered that progenitor cell hyperproliferation was linked to nonautonomous effects of Rac1 deletion on the macrophageal niche with heightened inflammation. Moreover, loss of Rac1 impaired cell death with autophagy but switched the cell death route to apoptosis. Finally, mammary gland reversibility failed in the absence of Rac1 as the alveoli failed to recommence lactation upon resuckling.

Funding: This work was supported by the Thomas, Berry and Simpson Fellowship (to NA), Wellcome Trust (#200588/Z/16/Z to NA) and Medical Research Council (MR/P028411/1 to NA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

We previously showed that the Rac1 GTPase plays a crucial role in postlactational mammary gland remodelling [ 9 ]. Rac1 is central to the conversion of MECs into nonprofessional phagocytes for the removal of dead cells and surplus milk and in controlling the inflammatory signature. We now reveal further roles for Rac1 in controlling the involution process. We have discovered that Rac1 acts as a nexus, controlling both the rate and balance of cell death and progenitor cell division in involution. Without Rac1, cell turnover accelerates with consequences on mammary gland remodelling in the irreversible phase. Moreover, failure to redifferentiate blocks mammary gland reversibility in the first phase of involution. Rac1 therefore has multifaceted roles in orchestrating the involution process.

Weaning of the infants triggers the mammary gland to enter postlactational involution, a process in which the surplus milk secreting alveolar epithelium is pruned from the ductal tree using a controlled cell death program [ 1 ]. In rodents, 90% of the alveolar epithelium laid down in pregnancy is subsequently removed during involution, with the balance of cell turnover tipped towards cell death and the remodelling process completed within approximately two weeks. In murine models of forced involution, simultaneous weaning of the pups at the peak of lactation causes the secretory alveoli to become engorged with milk as production continues for the first 24 h, after which they dedifferentiate. Both mechanical stretch and milk factors have been reported to stimulate cell death [ 1 , 2 ]. A number of programmed cell death mechanisms have been identified in postlactational involution, including apoptosis, lysosomal permeabilisation, and cell death with autophagy, although the significance of the different death pathways is unclear [ 3 – 7 ]. Moreover, lysosomal permeabilisation and autophagy may feed into an apoptotic death downstream. Involution occurs in two phases; In the first 48 h, cell death is triggered but the process of involution is reversible and the gland can reinitiate lactation once suckling resumes [ 3 , 8 ]. The second phase is irreversible and is characterised by destruction of the subtending basement membrane, extensive alveolar cell death, and repopulation of stromal adipocytes. The dead cells and residual milk are primarily removed by neighbouring live alveolar mammary epithelial cells (MECs) that act as phagocytes [ 9 – 12 ]. Engulfment of milk fat by the nonprofessional MECs triggers lysosomal permeabilisation through a stat-3-dependent mechanism, which ultimately kills the cells [ 12 ]. In the second phase, professional phagocytes from the immune system enter the gland to engulf the remaining dead cells and the tissue remodels back to a state closely resembling the nulliparous gland [ 13 , 14 ].

Cell turnover in adult tissues is characterised by the death of older cells and replacement with new through stem and progenitor cell proliferation. How these processes are balanced to maintain long-term tissue homeostasis is not clearly understood. The mammary gland is an example of a tissue that maintains a state of flux throughout the adult life. It also undergoes periods of profound growth and regression in each reproductive cycle, providing a tractable model to study cell turnover. The primary role of the mammary gland is to produce milk as a source of nutrients to feed the newborn. Successful lactation depends on the coordinated development and differentiation of the secretory alveolar epithelium during pregnancy and the subsequent removal of these milk-producing units once the milk supply is no longer needed. The balance of cell death and cell division continuously alters to permit tissue growth and regression within the mammary gland reproductive cycle, but little is known about how this is coordinated.

Results

Delayed alveolar regression and repopulation of adipocytes in involuting Rac1−/− mammary glands Mammary gland remodelling in involution is accompanied by alveolar regression and concomitant fat pad repopulation in the surrounding parenchyma. To investigate the role of Rac1 in tissue regression during involution, we examined mammary tissues from female mice triggered to involute through simultaneous weaning of the pups. The Rac1 gene was deleted specifically in luminal cells of the mammary gland using Rac1fl/fl:LSLYFP:WAPiCre (Rac1−/−) conditional knockout mice previously generated [15]. Cre-negative Rac1fl/fl:LSLYFP littermates were used as wildtype (WT) controls. Analysis of the WT mammary glands by histology and immunofluorescence staining of epithelial and adipocyte markers revealed significant lobular alveolar regression and adipocyte repopulation between involution days 2 to 4 (Figs 1A, 1C, 1E, 1G, 1I, 1J, 1K and S1). By contrast, alveoli in Rac1−/− glands remained distended with virtually no adipocyte repopulation (Figs 1B, 1D, 1F, 1H, 1K and S1). Four weeks postinvolution, the alveoli had completely regressed in both WT and Rac1−/− glands, confirming that lobular alveolar cell death had occurred in the absence of Rac1 (Fig 1L). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Delayed alveolar regression in Rac1−/− glands in involution. (A-D) Carmine staining of whole-mounted mammary glands of WT (A, C) and Rac1−/− (B, D) mice at postlactational involution day 2 and day 4. Note the alveolar regression in WT but not Rac1−/− glands at involution day 4. Bar: 5 mm, (inset 1 mm). (E-H) Haematoxylin–eosin (HE) stain shows alveolar regression and adipocyte repopulation in WT glands (E, G), but this is delayed in Rac1−/− glands (F, H). Bar: 100 μm. (I) Quantification of adipocyte repopulation in HE images. Error bars: +/− SEM of n = 3 mice per group. ** P ≤ 0.01. (J) Alveolar regression and adipocyte repopulation at involution day 4 were confirmed by immunofluorescence staining with WGA-488 (green) to detect epithelium and perilipin antibody (magenta) to detect adipocytes. Bar: 100 μm. (K) Quantification of alveolar regression and adipocyte repopulation. Error bars: +/− SEM of n = 3 mice per group. ** P ≤ 0.01. (J) Carmine staining of whole-mounted WT and Rac1−/− mammary glands at 4 weeks postlactational involution show complete regression of alveoli. Note: Bloated ducts persist in Rac1−/− glands. Bar: 0.7 mm. The data underlying the graphs shown in this figure can be found in S1 Data. https://doi.org/10.1371/journal.pbio.3001583.g001

β1-integrin is not upstream of Rac1 in alveolar regression In mammary gland tissue, β1-integrin functions upstream of Rac1 to regulate lactational differentiation, stem cell renewal, and cell cycle progression [9,16–19]; we thus investigated whether β1-integrin was linked to alveolar regression in involution. The β1-integrin gene was deleted in luminal cells of the mammary gland using β1-integrinfl/fl:LSLYFP:WAPiCre conditional knockout mice (Fig 2A). Crucially, tissue analysis at involution days 2 and 4 revealed that loss of β1-integrin did not impair alveolar regression and adipocyte repopulation compared to WT mice of a Cre-negative genotype (β1-integrinfl/fl:LSLYFP; Fig 2B–2F). This indicates that Rac1’s role in tissue regression is elicited through a distinct upstream signalling axis. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Ablation of β1-integrin does not phenocopy the Rac1−/− involution phenotype. (A) Ablation of β1-integrin in luminal cells at involution day 2 was detected by immunofluorescence staining with a β1-integrin antibody (red). YFP reporter gene expression (green) confirmed Cre-mediated recombination in transgenics. Bar: 50 μm. (B-E) HE stains at involution days 2 and 4 show no delay in alveolar regression in β1-integrin−/− mice. Both WT glands (A, B) and β1-integrin−/− (C, D) glands equally regressed. Bar: 100 μm. (E) Quantification of adipocyte repopulation. Error bars: +/− SEM of n = 4 mice per group. NS (not significant). The data underlying the graphs shown in this figure can be found in S1 Data. https://doi.org/10.1371/journal.pbio.3001583.g002

Rac1 ablation imbalances cell turnover rates in involution causing delayed alveolar regression We first investigated whether there was an initial delay in cell death in Rac1−/− glands, which might explain the delay in alveolar regression. At involution day 4 where the delayed regression is most prominent, numerous cell corpses were evident in Rac1−/− glands both within the alveolar epithelium and within the lumen space (S2A Fig). We confirmed the dead cells with cleaved caspase-3 staining and that they were of luminal cell origin with the Rosa:LSL-YFP reporter gene, which is activated in response to WAPiCre-induced recombination (Fig 3A and 3D). Moreover, cell death was not delayed within the earlier reversible involution stage (days 1 and 2), but rather increased cell corpses were detected in Rac1−/− tissue lumens (Figs 3B, 3C, 3E, 3F, S2A and S2B). However, this might be a result of defective clearance by epithelial phagocytes as opposed to increased cell death in Rac1−/− glands [9]. Thus, the delay in alveolar regression in transgenic glands is not linked to impaired cell death. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Delayed alveolar regression is not due to impaired cell death but heightened proliferation. (A-C) Immunofluorescence staining with cleaved caspase-3 (red) antibody shows numerous dead cells in Rac1−/− alveolar lumens at involution day 4 (A) and involution days 1 (B) and 2 (C). Keratin (Krt) 8/18 (green) was used to detect luminal cells. Note the Krt8-18 antibody cross-reacts with intact keratins in live cells and not cleaved forms in dead cells. Arrows: caspase-positive dead cells in the lumen. Bar: 40 μm. (D) Dead cells in the lumen were confirmed to be of luminal origin with the YFP reporter gene expression. GFP antibody was used to stain the YFP reporter gene. Arrowhead: dead cells in the lumen. Bar: 20 μm. (E, F) Quantification of cleaved caspase-3-positive cells at involution days 1 (E) and 2 (F). Error bars: +/− SEM of n = 3 mice per group. * P ≤ 0.05. (G, H) Immunofluorescence staining for proliferation marker Ki67 (red) reveals increased proliferation within Rac1−/− glands at involution day 4 in alveoli (G, G’) and ducts (H, H’). G’, H’ are zoomed images. Arrows point to ki67-positive luminal cells in alveoli and ducts. Epithelial tissue boundary was detected by smooth muscle actin (SMA, green) present in myoepithelial cells. Bar: 50 μm. (I) Proliferating cells were confirmed to be of luminal origin with the YFP reporter gene expression. Rac1wt:WAPiCre:YFP mice were used as WT controls to directly compare YFP-positive luminal cells. Ki67 antibody was used to detect proliferation. (J, K) Ki67 staining in mammary glands at day 2 and week 4 postlactational involution. Note: Proliferation persists in bloated Rac1−/− ducts at 4 weeks. Bar: 50 μm. (L-P) Quantitative analysis of Ki67 at involution day 4 (L, M), day 2 (N), day 1 (O), and week 4 (P). Proliferating cells were scored as either luminal or stromal/fat pad. SMA was used to count cells within the epithelial boundary (L, N-P). Luminal cell proliferation was confirmed by scoring YFP+/Ki67+ cells. Error bars: +/− SEM of n = 3 mice per group. *** P ≤ 0.001, ** P ≤ 0.01, ns; not significant. The data underlying the graphs shown in this figure can be found in S1 Data. https://doi.org/10.1371/journal.pbio.3001583.g003 Retention of milk within the lumens might contribute to the distended alveolar phenotype in Rac1−/− glands, since Rac1 is crucial for epithelial cell-directed engulfment of apoptotic cell corpses and residual milk [9]. However, given the extensive cell death detected in early involution, it was surprising that the alveoli remained intact. We thus investigated whether Rac1−/− alveoli were being maintained through cell renewal. Ki67 staining revealed almost no detectable proliferation in WT glands at involution day 4. In contrast, Rac1−/− mammary glands showed extensive proliferation within duct and alveolar luminal cells and within cells in the surrounding stromal/fat pad areas (Figs 3G, 3H, 3L and S2C). To confirm that proliferation was occurring within Rac1−/− luminal cells and not from cells that had escaped recombination, we scored proliferation in YFP-positive/Rac1−/− cells using the Rosa:LSL:YFP reporter gene (Fig 3I and 3M) Proliferation within the transgenic ducts continued at 4 weeks postinvolution; at this stage, however, most of the lobular alveoli had regressed (Figs 1L, 3K and 3P). Analysis of the phase I-reversible stage of involution (days 1 and 2) revealed no significant difference and very little proliferation at day 1. In contrast, at involution day 2, increased proliferation was detected in Rac1−/− tissues, but this was mainly confined to cells within the surrounding stromal/fat pad areas. Taken together, these results suggest that the delayed alveolar regression in early involution in Rac1−/− glands is linked to increased compensatory cell proliferation within the irreversible phase and not delayed cell death within the reversible phase. Thus, Rac1 has a key role in balancing the rate of cell death and proliferation in involution by limiting the division progeny of progenitors. Without Rac1, alveolar progenitors divide unexpectedly in involution. The newly replaced cells, however, have a limited life span and succumb to death as evidenced by subsequent alveolar regression 4 weeks postweaning involution.

Loss of Rac1 elicits distinct proliferation responses within the first and second gestation Rac1 has been linked to cell cycle progression in numerous cultured cells and in vivo tissues [19–22]. This is in complete contrast to our findings in vivo in the involuting mammary gland where loss of Rac1 induces proliferation. We thus examined the effects of Rac1 deletion on glandular development and proliferation within the first and second gestation. In the first gestation, histological analysis of mammary glands at pregnancy day 18 revealed slightly smaller alveoli in Rac1−/− glands, but the area occupied by adipocytes was not significantly different (Fig 4A–4C). Immunostaining with Ki67 and BrdU incorporation at lactation day 2 in the first cycle revealed no significant difference in proliferation between WT and Rac1−/− glands (Fig 4D–4G). Moreover, detection of the Rosa:LSL:YFP reporter gene revealed that recombination and thereby gene deletion was extensive and the proliferation was occurring within YFP-positive/Rac1−/− cells and not from WT cells that had escaped recombination (Fig 4D). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Distinct proliferation in Rac1−/− mammary glands within the first and second gestations. (A, B) Carmine staining of whole-mounted mammary glands (A) and HE stain (B) from WT and Rac1−/− mice at pregnancy day 18 in the first gestation. Bar: 2.8 mm (A) and 100 μm (B). (C) Quantification of adipocyte areas from HE stains show no significant difference. Error bars: +/− SEM of n = 4 mice per group. (D) Immunofluorescence staining with Ki67 in WT and Rac1−/− glands shows no difference in proliferation at lactation day 2 following the first gestation. Green fluorescent protein antibody was used to detect the WAPiCre-driven YFP reporter gene expression and hence Rac1 deletion. Arrows show proliferation in YFP-positive/Rac1−/−cells. Bar: 40 μm. (E) BrdU incorporation (red) to detect proliferating cells in WT and Rac1−/− mammary glands at lactation day 2, first cycle. WGA-488 (green) was used to demark mammary gland luminal cells. Bar: 40 μm. (F, G) Quantitative analysis of Ki67 staining (F) and BrdU (G) at lactation day 2 following the first gestation shows no significant difference between WT and Rac1−/− glands. Error bars: +/− SEM of n = 4 WT mice and n = 5 Rac1−/− mice. P > 0.05. (H, I) Carmine staining of whole-mounted mammary glands (H) and HE stain (I) from WT and Rac1−/− mice at pregnancy day 18 in the second gestation. Note the reduced lobular alveolar development. Bar: 2.8 mm (H) and 100 μm (I). (J) Quantification of adipocyte areas from HE stains show increased adipocyte area, thereby reduced alveologenesis in Rac1−/− mice. Error bars: +/− SEM of n = 4 mice per group. **P ≤ 0.01. (K, L) Ki67 staining (K) and BrdU incorporation (L) reveal reduced proliferation in Rac1−/− mammary glands at lactation day 2 following the second gestation. Bar: 40 μm. (M, N) Quantitative analysis of ki67 (M) and BrdU (N) positive staining in WT and Rac1−/− glands at lactation day 2, second gestation. Error bars: +/− SEM of n = 4 mice per group. **P ≤ 0.01. The data underlying the graphs shown in this figure can be found in S1 Data. https://doi.org/10.1371/journal.pbio.3001583.g004 In marked contrast to the involuting gland, within the second lactation cycle at day 2, there was a severe block in proliferation of luminal cells with concomitant reduced lobular alveolar development in both late pregnancy and early lactation stages (Figs 4H–4N and S3). Taken together, these data show that Rac1 deletion has no effect on proliferation within the first lactation, heightened proliferation in postlactational involution, and severely defective proliferation within the second lactation. The disparity in proliferation profiles suggests stage-specific cell autonomous and possible nonautonomous regulation by Rac1.

Cell death is accelerated without Rac1 We next investigated whether Rac1 affects the rate at which cells transit through death and whether cells die directly through primary or secondary necrosis. To test this, primary cultures of WT and Rac1−/− cells were induced to undergo anoikis, a detachment-induced cell death. We chose this method of cell death for three reasons; first, to prevent dying cells from removal by neighbouring nonprofessional phagocytosis, as single cells suspended in media are spatially out of reach for phagocytic removal; second, to trigger an innate programmed cell death rather than chemical-induced; and third, anoikis likely occurs in Rac1−/− glands in vivo as we previously showed loss of Rac1 perturbs cell-ECM adhesion with increased shedding of cells [9]. Dying cells incorporated higher levels of propidium iodide (PI) in the absence of Rac1 compared to WT controls, suggesting cell death by necrosis or late-stage apoptosis (Fig 7A–7C). To establish the proximal cell death route, we first examined for hallmarks of apoptosis as this process accompanies a series of well-defined biological steps. Both WT and Rac1−/− cell corpses displayed intact membranes with nuclear condensation, late-stage membrane blebbing, body fragmentation, and stained positive for cleaved caspase 3 indicative of an apoptotic cell death (Figs 3B, 3E and 7D–7G). Early-stage apoptosis is characterised by phosphatidylserine exposure to the outer membrane leaflet, and Annexin V is commonly used to detect this motif. Colabelling with Annexin V and PI in cells suspended for 1 h and 5 h revealed that approximately the same number of cells enter apoptosis with and without Rac1 (Annexin V only); however, by 5 h, significantly more Rac1−/− cells proceed to late-stage apoptosis/necrosis (Annexin V/ PI) with a concomitant reduction in numbers in early-stage apoptosis (Annexin V only; Fig 7H). Moreover, the numbers of viable cells declined following an 8-h suspension in the absence of Rac1 indicating that cells had proceeded through death and disintegrated within this time frame compared to WT controls (Fig 7I and 7J). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 7. Cell death proceeds faster without Rac1. (A-C) Increased PI uptake in Rac1−/− cells induced to die through anoikis in culture; (A) fluorescent image and (B) FACS analysis (C) Quantification of FACS. FC (fold change). Error bars: +/−SEM of n = 3 preps. **P ≤ 0.01. (D) Apoptotic blebs in WT and Rac1−/− dead cells in culture detected using EM. Arrows: membrane blebs. Bar: 2 μm. (E-G) EM images of (E) WT and (F, G) Rac1−/− tissues in vivo show apoptotic cells with nuclear pyknosis (E, F; arrows) and membrane blebbing (G). Double arrowhead: Apoptotic cells are engulfed by the alveolar epithelium in WT glands (E). Bar: (E, F) 2 μm, (G) 1 μm. (H) Annexin-V647+ and PI+ colabelled WT and Rac1−/− cells quantified by flow sorting following 1 h and 5 h in suspension show more cells proceed to late-stage apoptosis/necrosis without Rac1. (I) WT and Rac1−/− cells induced to die through anoikis for 8 h show reduced numbers of viable cells without Rac1. Cleaved caspase-3 was used to stain apoptotic cells. (J) Quantification of (I) showing total number of WT and Rac1−/− cells. Error bars: +/− SEM of n = 4. (K-N) EM images of WT and Rac1−/− in vivo tissues (K, L) and primary cultures (M, N) show cell necrosis without nuclear pyknosis and organelle release in Rac1−/− cells (L, N). In contrast, dying WT cells (K, M) have an intact cell membrane. Arrow: Nucleus released from necrotic cell without condensation suggests direct necrosis. Bar: (K) 2 μm, (L) 1 μm, (M, N) 0.5 μm. The data underlying the graphs shown in this figure can be found in S1 Data. https://doi.org/10.1371/journal.pbio.3001583.g007 As Annexin V can also bind necrotic cells with ruptured membranes, some of the cells in the AnnexinV/PI fraction may be a result of primary necrosis. To address whether necrotic cells occurred secondary to apoptosis as a result of defective phagocytosis or whether Rac1 loss triggered necrosis, we analysed the nuclei of necrotic cell corpses by electron microscopy. In Rac1−/− glands, the nuclei of ruptured cells were not condensed, suggesting they had not entered the apoptotic pathway first, but rather died through primary necrosis (Fig 7L). Organelle spillage and cell necrosis were also detected in Rac1−/− primary cultures (Fig 7N). In contrast, WT dead cells had intact membranes with nuclear condensation (Fig 7K and 7M). These studies reveal that Rac1 slows down the process of programmed cell death. Without Rac1, cells die either through primary necrosis or through apoptosis, but death proceeds more rapidly than in WT epithelia. Taken together, loss of Rac1 increases cell turnover rates in the involuting mammary gland through both increased progenitor proliferation and accelerated cell death.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001583

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.