Mammary Epithelial Cell Polarity Is Regulated Differentially by p73 Isoforms via Epithelial-to-mesenchymal Transition

Background: Both TAp73 and ΔNp73 are implicated in tumor suppression or promotion. However, the underlying mechanism remains unclear.

Results: Knockdown of p73, particularly TAp73 but not ΔNp73, disrupts cell polarity via up-regulation of EMT.

Conclusion: TAp73 is required for maintaining cell polarity by suppressing EMT.

Significance: This is the first study to explore the role of p73 isoforms in mammary epithelial cell mophorgenesis using a three-dimensional culture model.

Abstract

p73 is expressed as TA and ΔN isoforms, both of which are implicated in tumor suppression and/or promotion. To address how p73 possesses these opposing functions, we developed three-dimensional culture of MCF10A cells, which undergo cell morphogenesis to form polarized spheroids with hollow lumen similar to normal mammary acini in vivo. Here, we showed that upon knockdown of p73, particularly TAp73 but not ΔNp73, MCF10A cells formed irregular and near-normal acini without hollow lumen in three-dimensional culture. We also found that upon knockdown of p73 or TAp73, but not ΔNp73, MCF10A cells underwent epithelial-to-mesenchymal transition (EMT) via down-regulation of E-cadherin coupled with up-regulation of β-catenin and laminin V. In addition, we found that Snail-1, Slug, and Twist, all of which are known to act as EMT inducers by repressing E-cadherin expression, were increased markedly upon knockdown of p73 and TAp73 but little if any by ΔNp73. Furthermore, we showed that knockdown of p73 or TAp73 in MCF10A cells led to a marked increase in cell proliferation and migration. Together, our data suggest that TAp73 is necessary for maintaining normal cell polarity by suppressing EMT.

Introduction

p73, a member of the p53 family, is mapped at chromosome 1p36.2-3, a region where loss of heterozygosity occurs frequently in human cancer (1, 2). As a homolog of p53, p73 shares three major functional domains: the N-terminal transactivation domain, the central core sequence-specific DNA-binding domain, and the C-terminal oligomerization domain (1, 3, 4). p73 is expressed as TA and ΔN isoforms due to the use of alternative promoters, each of which consists of at least nine variants based on the splicing of the C-terminal domain (2). The biology and regulation of p73 is complex because p73 proteins possess both tumor-suppressing and -promoting functions (5–7). TAp73 generally is considered to have a property similar to wild-type p53 and capable of inducing p53 target genes that mediate prosurvival and prodeath activities (1, 4). However, ΔNp73 can be dominant-negative over TAp73 (8) and but also possess its own distinct transcriptional activity (9). As a result, ΔNp73 possesses a tumor-promoting activity under certain circumstances. Therefore, the balance of TAp73 versus ΔNp73 is considered as a prognostic factor correlated with the response of chemotherapy (2, 10).

In addition to its role in tumor suppression and promotion, p73 is known to play a role in development and differentiation of specific tissues and organs. For example, p73 is required for neuronal differentiation and development of central nervous and olfactory systems (11), and aberrant expression of p73 is involved in neurodegenerative diseases such as Alzheimer disease (12–14). Consistently, mice deficient in p73 show complex defects in neuronal development (7, 11, 15). p73 also is found to be highly expressed in airway ciliated columnar cells (10, 16) and in myoepithelial and basal cells of salivary gland (17, 18), breast (19), and prostate (19), highlighting a critical role for p73 in these tissues.

Mammary epithelial cells form polarized spheroid structures, also called acini, which consist of a central lumen, a single layer of polarized luminal epithelial cells surrounded by myoepithelial cells, and a basement membrane. Interestingly, disruption of the normal acinar architecture is a hallmark of mammary epithelial cell transformation (20). In the early stage of breast cancer, increased proliferation of epithelial cells is found coupled with a loss of acinar organization and filling of luminal space (20). To address how oncogenes or tumor suppress genes influence mammary epithelial cell polarity, three-dimensional culture of immortalized MCF10A cells, which form an acinar structure remarkably similar to the normal acinus in vivo, was established and used widely (21–26). To address how p73 is involved in tumor suppression and promotion via regulation of cell polarity, we generated MCF10A cell lines in which p73 or p73 isoforms were knocked down stably. We found that p73 and particularly TAp73 are necessary for maintaining normal cell polarity and suppressing epithelial-to-mesenchymal transition (EMT).³ Surprisingly, we found that knockdown of ΔNp73 alone has a modest effect on cell polarity and EMT. Thus, our study provides an insight into how p73 has two opposing functions in tumor suppression and promotion.

EXPERIMENTAL PROCEDURES

Cell Culture

The immortalized MCF10A cell line was obtained from American Type Culture Collection (ATCC, Manassas, VA) and cultured as described previously (25). The overlay three-dimensional culture was carried out as described previously with some modifications (25, 27). Briefly, four-well chamber slides (Millipore Corp., Danvers, MA) were pre-coated evenly with 80 μl of overnight-thawed Matrigel, and MCF10A cells were plated onto Matrigel-coated chamber slides at 5,000 cells/well in complete growth medium with 2% Matrigel and allowed to grow for 1–20 days. Overlay medium containing 2% Matrigel was renewed every 4 days.

Reagents

Growth factor-reduced Matrigel was purchased from BD Transduction Laboratories. DMEM/F12 medium, donor horse serum, TO-PRO-3 nucleus dye, and anti-mouse antibody conjugated to fluorophore 488 were purchased from Invitrogen. Hydrocortisone, insulin and cholera toxin were purchased from Sigma. Recombinant human epidermal growth factor (EGF) was purchased from Peprotech (Rocky Hill, NJ). Normal goat serum was purchased from Jackson ImmunoResearch Laboratories (West Grove, PA).

Plasmid Construction and Cell Line Generation

To generate stable shRNA against p73, ΔNp73, TAp73 under the control of the U6 promoter, two 62-base oligonucleotides were annealed and then cloned into pBabe-U6 shRNA expression vector, and the resulting plasmids were designated as pBabe-U6-shp73, pBabe-U6-shΔNp73, and pBabe-U6-shTAp73. The shRNA oligonucleotides used are listed with the siRNA targeting region shown in uppercase letters (Table 1). The resulting knockdown cell lines were selected with puromycin and confirmed by RT-PCR and/or Western blot analysis.

TABLE 1.

The oligonucleotides used for generation of shRNA expression vectors

p73 shRNA1	Sense, 5′-tcgaggtccGCTGATGAGGACCACTACCttcaagagaGGTAGTGGTCCTCATCAGCtttttg-3′; antisense, 5′-gatccaaaaaGCTGATGAGGACCACTACCtctcttgaaGGTAGTGGTCCTCATCAGCggacc-3′
p73 shRNA2	Sense, 5′-tcgaggtccCCGCTCTTGAAGAAACTCTttcaagagaAGAGTTTCTTCAAGAGCGGtttttg-3′; antisense, 5′-gatccaaaaaCCGCTCTTGAAGAAACTCTtctcttgaaAGAGTTTCTTCAAGAGCGGggacc-3′
TAp73 shRNA1	Sense, 5′-tcgaggtccGGCATGACTACATCTGTCAttcaagagaTGACAGATGTAGTCATGCCtttttg-3′; antisense, 5′-gatccaaaaaGGCATGACTACATCTGTCAtctcttgaaTGACAGATGTAGTCATGCCggacc-3′
TAp73 shRNA2	Sense, 5′-tcgaggtccCCAGACAGCACCTACTTCGttcaagagaCGAAGTAGGTGCTGTCTGGtttttg-3′; antisense, 5′-gatccaaaaaCCAGACAGCACCTACTTCGtctcttgaaCGAAGTAGGTGCTGTCTGGggacc-3′
ΔNp73 shRNA1	Sense, 5′-tcgaggtccGACAGAACTAAGGGAGATGttcaagagaCATCTCCCTTAGTTCTGTCtttttg-3′; antisense, 5′-gatccaaaaaGACAGAACTAAGGGAGATGtctcttgaaCATCTCCCTTAGTTCTGTCggacc-3′
ΔNp73 shRNA2	Sense, 5′-tcgaggtccGGATTCAGCCAGTTGACAGttcaagagaCTGTCAACTGGCTGAATCCtttttg-3′; antisense, 5′-gatccaaaaaGGATTCAGCCAGTTGACAGtctcttgaaCTGTCAACTGGCTGAATCCggacc-3′

RT-PCR Analysis

Total RNA was extracted from cells using TRIzol (Invitrogen) according to the manufacturer's instructions. cDNA was synthesized using M-MLV reverse transcriptase kit (Promega Corp.) according to the manufacturer's protocol. The mRNA levels of p73 and its isoforms was measured by PCR with p73 and p73 isoform-specific primers. The primers to detect p73 are 5′-cagcagcagcagctcctaca-3′ (sense) and 5′-cccaggtcctgacgaggct-3′ (antisense). The primers to detect ΔNp73 are 5′-gatccatgccctcgtcccac-3′ (sense) and 5′-ctgctcatctggtccatgg-3′ (antisense). The primers to detect TAp73 were 5′-cagacagcacctacttcgac-3′ (sense) and 5′-ctgctcatctggtccatgg-3′ (antisense). The actin gene was chosen as a loading control and detected with primers 5′-ctgaagtaccccatcgagcacggca-3′ (sense) and 5′-ggatagcacagcctggatagcaacg-3′ (antisense).

Western Blot Analysis

Western blot was performed as described (28). Antibodies used were purchased from Calbiochem (anti-p73 (ab-1)-ER-13, anti-p73 (ab-2)-ER-15), Millipore (anti-laminin γ2), Santa Cruz Biotechnology (anti-β-catenin (E-5), anti-Snail-1, anti-Twist, p21, and anti-GAPDH), Cell Signaling (anti-Slug), BD Transduction Laboratories (anti-E-cadherin), Prosci (PUMA), Sigma (anti-actin), and Bio-Rad (secondary antibodies against rabbit or mouse IgG conjugated with HRP).

Confocal Microscopy

The three-dimensional structures in Matrigel were fixed in 4% paraformaldehyde at room temperature for 20 min and permeabilized with 0.5% Triton X-100 in PBS for 30 min at 4 °C. After quenching with 100 mm glycine in PBS, three-dimensional structures were pre-blocked in a primary blocking buffer A (130 mm NaCl, 7 mm Na₂HPO₄, 3.5 mm NaH₂PO₄, 0.1% BSA, 0.2% Triton X-100, and 0.05% Tween 20) containing 10% normal goat serum for 2 h and further blocked in a secondary blocking buffer B (buffer A plus 10% normal goat serum and 20 mg/ml of goat anti-mouse F(ab′)₂ fragments) for 1 h. Three-dimensional cultures were incubated overnight with primary antibodies at 4 °C, washed thoroughly with buffer A with gentle shaking, and stained with FITC-conjugated secondary antibody (diluted 1:200 in buffer A containing 10% goat serum) for 1 h. The structures were nuclear stained with 5 μg/ml of TO-PRO-3 in PBS for 15 min at room temperature. The TO-PRO-3 stain was removed by washing the chamber slide with PBS for 5 min, and then the slide was mounted under a glass coverslip with 0.1% para-phenylenediamine D (PPD) and 90% glycerol in PBS. Microscopic analysis was performed using a confocal microscopy system (Axiovert 100m, Zeiss) and images were acquired using the software for Carl Zeiss laser scanning microscope (LSM-510). The images of acinar structures were captured by the Z-stacking function for serial confocal sectioning at 2-μm intervals. Experiments were conducted in triplicate.

Colony Formation Assay

MCF10A cells were cultured in a six-well plate for ∼12 days and then fixed with methanol/glacial acetic acid (7:1) followed by staining with 0.1% crystal violet.

In Vitro Cell Migration Assay

For wound healing assay, cells were grown in a six-well plate for 24 h. The monolayers were wounded by scraping with a P200 micropipette tip and washed two times with PBS. At specified time points after the scraping, cell monolayers were photographed using a Canon EOS 40D digital camera (Canon, Lake Success, NY). Cell migration was determined by visual assessment of cells migrating into the wound using phase contrast microscopy.

Statistical Analysis

Data were presented as mean ± S.D. Statistical significance was determined by Student's t test. Values of p < 0.05 were considered significant.

RESULTS

p73 Is Required for Normal Acinar Formation

The spontaneously immortalized and non-tumorigenic MCF10A epithelial cells possess three-dimensional growth ability and form a polarized structure surrounding a hollow lumen. In this process, proapoptotic genes or tumor suppressor genes promote, whereas oncogenes block acinar differentiation (29–32). Here, we showed that in three-dimensional culture, parental MCF10A cells formed normal cobblestone-like epithelial cell morphology in two-dimensional culture (supplemental Fig. S1, A, a panel) and acinus-like structures in three-dimensional culture (supplemental Fig. S1, A, b and c panels) along with hollow lumen (supplemental Fig. S1, C–E), consistent with our and other previous reports (21–25). In addition, cells showed an apical-basal distribution of polarity marker laminin V and cell-cell junction markers, such as E-cadherin and β-catenin (supplemental Fig. S1, C–E). To explore the role of p73 in MCF10A cell morphogenesis, we generated multiple MCF10A cell lines in which endogenous p73 was stably knocked down by shRNA. RT-PCR was performed and showed that the level of transcripts for p73, TAp73, and ΔNp73 was decreased (Fig. 1A, clones 1 and 2). Consistent with this, Western blot analysis showed that the level of TAp73α and ΔNp73α proteins was also decreased (Fig. 1B, clones 1 and 2). Because the α-isoform, which is the predominant form of TAp73 and ΔNp73, can be detected by Western blotting, the levels of TAp73α and ΔNp73α were measured and used to represent relative expression levels of TAp73 and ΔNp73, respectively. Interestingly, we found that upon knockdown of p73 (p73-KD), MCF10A cells exhibited spindle-like morphology in two-dimensional culture (Fig. 1C, a panel), and both irregular and near-normal acinus-like structures in three-dimensional culture (Fig. 1C, b and c panels). We also found that the lumen of the acini was not cleared, and E-cadherin staining at cell-cell junctions was decreased markedly in p73-KD cells (Fig. 1D and supplemental Fig. S2A). In addition, polarized lateral distribution of β-catenin and apical-basal distribution of laminin V were disrupted as β-catenin and laminin V were found to be expressed in the nuclear and cytosol of cells in the filled lumen (Fig. 1, E and F, and supplemental Fig. S2, B and C). These results suggest that p73 is necessary for the centrally localized cells to undergo apoptosis for lumen formation and thus proper organization of acini.

FIGURE 1.

p73 is required for acinar formation and maintenance of cell polarity and cell-cell junction in three-dimensional culture. A and B, generation of MCF10A cells in which endogenous p73 was stably knocked down (clones 1 and 2). The levels of p73, TAp73, and ΔNp73 mRNAs were determined by RT-PCR (A). The levels of TAp73α and ΔNp73α proteins in MCF10A and p73-KD cells were measured by Western blot (B). The blots were probed with antibodies against TAp73, ΔNp73, and actin, respectively. C, representative phase-contrast microscopic images of MCF10A cells with p73-KD in two-dimensional culture (a panel, 200×) and three-dimensional culture (b panel, 40×; c panel, 100×). A black arrow indicates elongated spindle-like MCF10A cells. D, representative confocal images of cross-sections through the middle of an acinus stained with TO-PRO-3 and antibody against E-cadherin (all 250×). Scale bar, 20 μm. E, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against β-catenin. White arrows in E indicate the accumulation and translocation of β-catenin in an acinar structure. F, representative confocal images of cross-sections through the middle of an acinus stained with TO-PRO-3 and antibody against laminin V.

TAp73 but Not ΔNp73 Is Required for Formation of Mammary Epithelial Acini

Because p73 is expressed as TA and ΔN isoforms, we wanted to determine the role of each p73 isoform in morphogenesis of MCF10A cells. To test this, we generated multiple MCF10A cell lines in which ΔN but not TA isoform was knocked down by shRNA (Fig. 2, A and B, clones 8 and 17). We found that ΔNp73-KD cells exhibited near-normal cell morphology in both two-dimensional (Fig. 2C, a panel) and three-dimensional (Fig. 2C, b and c panels) cultures. In addition, the distribution of tight junction markers, E-cadherin and β-catenin, was retained (Fig. 2, D and E, and supplemental Fig. S3, A and B) along with basal deposition of laminin V (Fig. 2F and supplemental Fig. S3C).

FIGURE 2.

ΔNp73 is not required for formation of mammary epithelial acini. A and B, generation of MCF10A cells in which ΔNp73 was stably knocked down (clone 8 and 17). The levels of ΔNp73, p73, and TAp73 transcripts were determined by RT-PCR (A). The levels of ΔNp73α and TAp73α proteins were measured by Western blotting (B). C, representative images of MCF10A cells with ΔNp73-KD in two-dimensional culture (a panel, 200×) and three-dimensional culture (b panel, 40×; c panel, 100×). D, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against E-cadherin in MCF10A cells with ΔNp73-KD (all 250×). Scale bar, 20 μm. E, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against β-catenin in MCF10A cells with ΔNp73-KD. F, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against laminin V in MCF10A cells with ΔNp73-KD.

Next, we generated multiple MCF10A cell lines in which TA but not ΔN isoform was stably knocked down by shRNA (Fig. 3, A and B, clones 1 and 2). In contrast to ΔNp73-KD, TAp73-KD altered cell morphology in two-dimensional culture (Fig. 3C, a panel) and led to formation of abnormal and near-normal acini in three-dimensional culture (Fig. 3C, b and c panels). In addition, the lumen of the acini was not cleared and E-cadherin staining at cell-cell junctions was significantly decreased (Fig. 3D and supplemental Fig. S4A). Furthermore, β-catenin and laminin V were found to be expressed in the nuclear and cytosol of cells in the filled lumen (Fig. 3, E and F, and supplemental Fig. S4, B and C). This suggests that the role of TAp73 in cell polarity recapitulates that for p73.

FIGURE 3.

TAp73 is required for formation of mammary epithelial acini. A and B, generation of MCF10A cells in which TAp73 was stably knocked down (clone 1 and 2). The levels of ΔNp73, p73, and TAp73 transcripts were determined by RT-PCR (A). The levels of ΔNp73α and TAp73α proteins were measured by Western blotting (B). C, representative images of MCF10A cells with TAp73-KD in two-dimensional culture (a panel, 200×) and three-dimensional culture (b panel, 40×; c panel, 100×). A black arrow indicates elongated spindle-liked MCF10A cells. D, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against E-cadherin in MCF10A cells with TAp73-KD (All 250×). Scale bar, 20 μm. E, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against β-catenin in MCF10A cells with TAp73-KD. White arrows indicate the accumulation and translocation of β-catenin in an acinar structure. F, representative confocal images of cross-sections through the middle of acini stained with TO-PRO-3 and antibody against laminin V in MCF10A cells with TAp73-KD.

p73 Knockdown Alters Cell Polarity via EMT

Down-regulation of E-cadherin coupled with up-regulation of β-catenin and laminin V is considered a hallmark of EMT (33). Thus, we examined whether knockdown of p73 alters the expression pattern of these and other EMT markers in MCF10A cells cultured in Matrigel for 20 days. Indeed, we found that upon knockdown of p73 or TAp73, the level of E-cadherin was decreased (Fig. 4A), whereas the levels of laminin V and β-catenin were increased (Fig. 4B), consistent with altered staining patterns and intensities of these EMT markers in acinus-like structures (Figs. 1–3, D–F). In contrast, ΔNp73-KD slightly increased the level of laminin V but had little effect on the level of E-cadherin and β-catenin in MCF10A cells (Fig. 4, A and B, compare lane 1 versus 3). In addition, we found that Snail-1, Slug, and Twist, all of which are known to function as EMT inducers by repressing E-cadherin expression (33–36), were increased markedly upon knockdown of p73 and TAp73 but little if any by ΔNp73 (Fig. 4B, compare lanes 2–4 with 1). Furthermore, we found that the levels of p21 and PUMA in MCF10A cells were decreased upon knockdown of p73 and TAp73 but modestly increased upon knockdown of ΔNp73 (Fig. 4C, compare lanes 2–4 with 1). This result is consistent with previous reports that p21 and PUMA are highly induced by, and serve as a mediator of, TAp73 but not ΔNp73 in growth suppression (37–40). Next, colony formation and wound healing assays were performed to measure cell proliferation and migration, both of which are characteristics of EMT (41). We showed that upon knockdown of p73 and TAp73, the ability of MCF10A cells to proliferate and to migrate was increased markedly (Fig. 5, A and B). However, ΔNp73-KD decreased cell proliferation (Fig. 5A) but had little if any on cell migration (Fig. 5B). Together, these findings demonstrated that knockdown of p73 and in particular TA isoform alters cell polarity via EMT.

FIGURE 4.

p73-KD and TAp73-KD, but not ΔNp73-KD induce EMT phenotypes in MCF10A cells. Western blots were prepared using extracts from MCF10A cells (lane 1), and MCF10A cells with p73-KD (lane 2), with ΔNp73-KD (lane 3), or with TAp73-KD (lane 4). MCF10A cells were grown in Matrigel for 20 days. The blots were probed with antibodies against E-cadherin (A), β-catenin (B), laminin V (B), Snail (B), Slug (B), Twist (B), p21 (C), PUMA (C), and actin (A–C), respectively. The basal levels of each gene were arbitrarily set at 1.0, and fold change is shown below each lane.

FIGURE 5.

p73-KD and TAp73-KD, but not ΔNp73-KD, promote cell proliferation and migration in MCF10A cells. A, top panel: colony formation assay was performed with MCF10A cells, or MCF10A cells with p73-KD, with ΔNp73-KD or with TAp73-KD. Cells were cultured for a period of 12 days and then fixed and stained with crystal violet. Bottom panel: the number of colonies was counted and presented as mean ± S.D. from three separate experiments. B, wound healing assay was performed with MCF10A cells, or MCF10A cells with p73-KD, with ΔNp73-KD or with TAp73-KD. Cell migration was determined by visual assessment of cells migrating into the wound for a period of 24 h using a phase-contrast microscopy. C, a model for the role of p73 in cell polarity.

DISCUSSION

p73 is known to play a role in tumor suppression and promotion as well as development and differentiation of specific tissues and organs. Because TAp73 and ΔNp73 often possess opposing functions, it is not clear which isoform is involved in these processes. It is well known that in three-dimensional culture, normal mammary epithelial cells form polarized, spherical acini with hollow lumen whereas tumor mammary epithelial cells form large, nonpolarized, undifferentiated aggregates without lumen (42). Thus, we take the advantage of MCF10A three-dimensional culture model to examine the role of p73 isoforms in the process of mammary epithelial cell morphogenesis. First, we found that knockdown of p73, particularly TAp73 disrupts, whereas knockdown of ΔNp73 has limited effect on, the MCF10A acinar structure, suggesting that TAp73 is required for MCF10A cells to form polarized acinar structures with hollow lumen. Second, cell polarity is altered by knockdown of TAp73 at least in part via induction of EMT since the expression pattern of EMT markers (laminin V, E-cadherin, β-catenin, Snail-1, Slug, and Twist) are altered by knockdown of TAp73 along with increased cell proliferation and migration, whereas knockdown of ΔNp73 has limited effect on EMT regulation. Taken together, our data suggest that TAp73 maintains normal cell polarity by suppressing EMT, whereas ΔNp73 promotes cell proliferation but has little if any effect on normal cell morphogenesis (Fig. 5C).

It is well established that the primary event in acinar formation is the establishment of epithelial cell polarity, which then modulates cell proliferation and cell death required for acinus maturation and lumen formation (22). Here, we found that TAp73 and ΔNp73, both of which are expressed in MCF10A cells, differentially regulate mammary epithelial cell polarity and gene expression. Specifically, we showed that knockdown of p73 or TAp73 leads to disruption of acinar formation and lumen clearance, indicating that TAp73 is the major isoform to regulate genes that are required for lumen clearance. Indeed, we found that p73 or TAp73 knockdown decreased the expression of p73 target genes, including p21 and PUMA, whereas knockdown of ΔNp73 leads to a modest increased expression of p21 and PUMA (Fig. 4C). Considering that p21 mediates p73-dependent cell cycle arrest and that PUMA mediates p73-dependent apoptosis, our observations are consistent with the postulation that both anti-proliferative and apoptotic activities are required for achieving lumen formation in mammary epithelial acini (22). Thus, further study is warranted to address whether p21 and PUMA mediate cell polarity and lumen formation in MCF10A three-dimensional culture.

The EMT process is characterized by down-regulation of epithelial markers and acquisition of mesenchymal properties, including loss of E-cadherin, a prerequisite for epithelial cells to acquire migratory properties (43). As a result, EMT promotes cells to dissociate from epithelial tissues and become more motile (44), which occurs in many pathological situations, such as wound healing, fibrosis, and epithelial tumor development. In this study, we found that knockdown of p73 or TAp73 disrupts cell polarity, accompanied by loss of E-cadherin, accumulation of β-catenin in the nucleus, and increased expression of laminin V. Consistent with the notion that E-cadherin is repressed by Snail-1, Slug, and Twist (33, 45), we found that Snail-1, Slug, and Twist were up-regulated by knockdown of p73 or TAp73. The alteration of EMT markers indicates that upon knockdown of p73 or TAp73, MCF10A cells undergo EMT. In line with this, we observed that knockdown of p73 or TAp73 leads to formation of acini with filled lumen and acquisition of enhanced migratory activity.

In summary, this study provides evidence that only TAp73 is required for proper mammary epithelial acinar formation and for suppressing cell migration and invasion by regulating expression of E-cadherin and other EMT markers (Fig. 5C). In contrast, ΔNp73 is required for proper cell proliferation. These results highlight the critical role of the p73-EMT pathway in the mammary epithelial cell polarity, which might be explored for development of novel therapeutic strategies for breast cancer and other malignances.