CAPER, a novel regulator of human breast cancer progression

Abstract

CAPER is an estrogen receptor (ER) co-activator that was recently shown to be involved in human breast cancer pathogenesis. Indeed, we reported increased expression of CAPER in human breast cancer specimens. We demonstrated that CAPER was undetectable or expressed at relatively low levels in normal breast tissue and assumed a cytoplasmic distribution. In contrast, CAPER was expressed at higher levels in ductal carcinoma in situ (DCIS) and invasive ductal carcinoma (IDC) specimens, where it assumed a predominantly nuclear distribution. However, the functional role of CAPER in human breast cancer initiation and progression remained unknown. Here, we used a lentiviral-mediated gene silencing approach to reduce the expression of CAPER in the ER-positive human breast cancer cell line MCF-7. The proliferation and tumorigenicity of MCF-7 cells stably expressing control or human CAPER shRNAs was then determined via both in vitro and in vivo experiments. Knockdown of CAPER expression significantly reduced the proliferation of MCF-7 cells in vitro. Importantly, nude mice injected with MCF-7 cells harboring CAPER shRNAs developed smaller tumors than mice injected with MCF-7 cells harboring control shRNAs. Mechanistically, tumors derived from mice injected with MCF-7 cells harboring CAPER shRNAs displayed reduced expression of the cell cycle regulators PCNA, MCM7, and cyclin D1, and the protein synthesis marker 4EBP1. In conclusion, knockdown of CAPER expression markedly reduced human breast cancer cell proliferation in both in vitro and in vivo settings. Mechanistically, knockdown of CAPER abrogated the activity of proliferative and protein synthesis pathways.

Breast cancer is one of the most common cancers in women and is a major cause of death worldwide. Estrogens stimulate mammary epithelial cell proliferation and contribute to the development and progression of human breast cancer.^1-3 In fact, prolonged exposure to estrogens (from early menarche and/or late menopause) is a well-known risk factor for breast cancer development.^4,5 Estrogens bind members of the nuclear receptor superfamily known as estrogen receptor (ER)α and ERβ.^6-8 Ligand-occupied ERs then directly interact with DNA at the estrogen response element (ERE) to activate or repress the transcription of target genes.^9,10 Approximately 60–70% of human breast cancer cases express ERα and depend on estrogen for their growth.^11,12 Accordingly, anti-estrogen-related therapies are widely use for the treatment of ER-positive human breast cancer.^2,3,13 However, 40–50% of ER-positive breast cancer patients fail to respond to anti-estrogen therapies.^14-17 Importantly, the transcriptional activity of nuclear receptors, such as ER, is tightly regulated by their interactions with co-regulators.^9,18 In fact, the recruitment of co-activators and co-repressors to the transcriptional complex is essential for efficient transcriptional control.^9,18 Thus, modulation of ER co-regulator expression could disturb ER transcriptional activity and affect the progression of ER-positive human breast cancers.

CAPER (co-activator of AP-1 and ER), also known as RNA-binding protein-39 (Rbm39) and hepatocellular carcinoma-1.4 (HCC1.4), is a co-activator of ERα, ERβ, and the activator protein-1 (AP-1) component c-Jun that couples transcription with pre-mRNA processing.¹⁹ Using glutathione-S-transferase (GST)-pulldown and luciferase reporter assays, Jung et al. (2002) demonstrated that CAPER directly interacted with ERα/β and c-Jun and potently stimulated their transcriptional activity in vitro.¹⁹ However, the role of CAPER in human breast cancer pathogenesis remained unknown. Interestingly, using the PM2000 AQUA automated quantitative analysis system (HistoRx Inc) and a human breast cancer tissue microarray consisting of 84 invasive ductal carcinomas (IDCs) and 20 normal breast tissue samples, we recently reported that CAPER protein expression was significantly elevated in human breast cancer specimens, as compared with normal breast tissue.²⁰ Our immunohistochemical analyses further revealed that CAPER was undetectable or expressed at relatively low levels in normal breast tissue and assumed a cytoplasmic distribution.²⁰ In contrast, CAPER was expressed at higher levels in ductal carcinoma in situ (DCIS) and IDC specimens, where it assumed a predominantly nuclear distribution.²⁰ However, the functional role of CAPER in human breast cancer initiation and progression remained elusive.

Here, we used a lentiviral-mediated gene silencing approach to reduce the expression of CAPER in MCF-7 cells, an ER-positive human breast cancer cell line. The proliferation and tumorigenicity of MCF-7 cells stably expressing control or CAPER shRNAs was then evaluated in both in vitro and in vivo experiments.

Results

Knockdown of CAPER expression reduces the proliferation of MCF-7 cells in vitro

To gain insight into the role of CAPER in the proliferation of ER-positive human breast cancer cells, MCF-7 cells stably expressing either control or human CAPER shRNAs were seeded at equal density in 12-well culture dishes. Seven days later, cells were trypsinized and counted, and the number of cells expressed as a percentage of the control group. As shown in Figure 1A, MCF-7 cells stably expressing human CAPER shRNAs displayed reduced CAPER protein levels. Importantly, knockdown of CAPER expression significantly reduced the proliferation (~4-fold) of MCF-7 cells in vitro (P < 0.001, n = 4 for each group, Fig. 1B).

Figure 1. Knockdown of CAPER expression reduces the proliferation of MCF-7 cells in vitro. (A) Western blot analysis shows reduced CAPER protein levels in MCF-7 cells stably expressing CAPER shRNAs. Immunoblot analysis with GAPDH is shown as a control for equal loading. (B) Knockdown of CAPER expression markedly reduced the proliferation (~4-fold) of MCF-7 cells in vitro (*P < 0.001 vs control shRNA, n = 4).

Knockdown of CAPER expression markedly reduces tumor growth in vivo

To determine the functional role of CAPER in human breast cancer initiation and progression, MCF-7 cells harboring control or CAPER shRNAs were injected into the right mammary fat pad of female nude mice supplemented with 17β-estradiol. Six weeks post-injection, mice were euthanized by CO₂ inhalation, and tumors were harvested, weighed and measured. As shown in Figure 2A, mice injected with MCF-7 cells stably expressing CAPER shRNAs developed smaller tumors than those injected with MCF-7 cells harboring control shRNAs. In fact, quantitative analyses demonstrated a ~2-fold reduction in tumor weight (P < 0.001, Fig. 2B) and ~3-fold reduction in tumor volume (P < 0.001, Fig. 2C) in mice injected with MCF-7 cells harboring CAPER shRNAs (n = 14–20 for each group). Strikingly, when tumors were stratified as either small (≤0.1 g) or large (>0.1 g) tumors, only 21.4% of mice injected with MCF-7 cells harboring CAPER shRNAs displayed large tumors as compared with 88.0% of mice injected with MCF-7 cells harboring control shRNAs (Fig. 2D).

Figure 2. Knockdown of CAPER expression markedly reduces tumor growth in vivo. (A) Mice injected with MCF-7 cells harboring CAPER shRNAs developed smaller tumors (indicated by arrows) than mice injected with MCF-7 cells harboring control shRNAs. Quantitative analyses demonstrated a ~2-fold reduction in tumor weight (B, *P < 0.001 vs. control shRNA) and ~3-fold reduction in tumor volume (C, *P < 0.001 vs control shRNA) in mice injected with MCF-7 cells harboring CAPER shRNAs (n = 14–20 for each group). (D) When tumors were stratified as either small (≤0.1 g) or large (>0.1 g), 21.4% of mice injected with human breast cancer cells harboring CAPER shRNAs displayed large tumors as compared with 88.0% of mice injected with cells harboring control shRNAs.

CAPER expression positively correlates with tumor size

In order to determine if the expression of CAPER correlated with tumor size, we performed Pearson correlation analyses using the GraphPad InStat software. Interestingly, as shown in Figure 3, CAPER protein levels positively correlated with both the tumor weight (P = 0.0002, n = 31, Fig. 3A) and tumor volume (P = 0.0042, n = 31, Fig. 3B).

Figure 3. The expression of CAPER positively correlates with tumor size. Pearson correlation analyses demonstrate that endogenous CAPER expression positively correlates with both the (A) tumor weight (P = 0.0002, n = 31) and (B) tumor volume (P = 0.0042, n = 31). Tumors derived from mice injected with MCF-7 cells harboring control shRNAs are represented as black squares (n = 20), while tumors derived from mice injected with MCF-7 cells harboring CAPER shRNAs are represented as red squares (n = 11).

Knockdown of CAPER expression modulates the expression of cell cycle regulators

In order to gain further mechanistic insight, tumors collected from nude mice injected with MCF-7 cells harboring either control or CAPER shRNAs were cut in half, and each half was either frozen in liquid nitrogen or fixed in 4% paraformaldehyde (PFA). The expression of proliferation markers was then determined by both immunoblot and immunohistochemical analyses. Interestingly, our immunoblot analyses demonstrate that tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs display reduced expression of positive regulators of cell cycle progression, such as the proliferating cell nuclear antigen (PCNA), mini-chromosome maintenance protein-7 (MCM7), and Cyclin D1 (Fig. 4A). Importantly, as shown in Figure 4B, our immunohistochemical analyses support our immunoblot data and demonstrate a reduction of PCNA, MCM7, and Cyclin D1 expression in tumors derived from mice injected with MCF-7 cells stably expressing CAPER shRNAs.

Figure 4. Knockdown of CAPER expression modulates the expression of cell cycle regulators. (A) Immunoblot analyses demonstrate that tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs display reduced expression of the cell cycle regulators PCNA, MCM7, and Cyclin D1. Immunoblot analysis with GAPDH is shown as a control for equal loading. (B) Immunohistochemical analyses confirm the reduced expression of PCNA, MCM7, and Cyclin D1 in tumors derived from mice injected with MCF-7 cells stably expressing CAPER shRNAs.

Knockdown of CAPER expression abrogates the protein synthesis pathway

The expression of the protein synthesis marker eukaryotic initiation factor 4E-binding protein-1 (4EBP1) was also determined by both western blot and immunohistochemical analyses. Interestingly, our immunoblot analyses demonstrate that tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs display reduced levels of phospho-4EBP1 (Fig. 5A). As shown in Figure 5B, our immunohistochemical analyses also support our immunoblot data and demonstrate a reduction of phospho-4EBP1 levels in tumors derived from mice injected with MCF-7 cells harboring CAPER shRNAs.

Figure 5. Knockdown of CAPER expression reduces the levels of protein synthesis markers. (A) Immunoblot analyses demonstrate that tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs display reduced levels of the protein synthesis marker phospho-4EBP1. Immunoblot analysis with GAPDH is shown as a control for equal loading. (B) Immunohistochemical analysis confirms the reduced levels of phospho-4EBP1 in tumors derived from mice injected with MCF-7 cells stably expressing CAPER shRNAs.

Knockdown of CAPER expression attenuates the phosphorylation of c-Jun

The phosphorylation of c-Jun was determined by immunoblot and immunohistochemical analyses. As shown in Figure 6A, our immunoblot analyses demonstrate that tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs display reduced levels of both phospho(ser63)-c-Jun and phospho(ser73)-c-Jun. Accordingly, our immunohistochemical analyses also revealed decreased phospho(ser63)-c-Jun and phospho(ser73)-c-Jun levels in tumors stably expressing CAPER shRNAs (Fig. 6B).

Figure 6. Knockdown of CAPER expression reduces the phosphorylation of c-Jun. (A) Immunoblot analyses demonstrate that tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs display reduced levels of phospho(ser63)-c-Jun and phospho(ser73)-c-Jun. Immunoblot analysis with GAPDH is shown as a control for equal loading. (B) Immunohistochemical analyses confirm the reduced levels of phospho(ser63)-c-Jun and phospho(ser73)-c-Jun in tumors derived from mice injected with MCF-7 cells stably expressing CAPER shRNAs.

Discussion

CAPER was first identified based on its interaction with the activating signal co-integrator-2 (ASC-2), another nuclear receptor co-activator.¹⁹ CAPER is identical to the human protein HCC1.4, which was previously described as a novel nuclear auto-antigen in human hepatocarcinoma.^19,21 CAPER was also shown to be related to known splicing factors such as the U2 auxiliary factor (U2AF65) and the poly(U)-binding-splicing factor-60 (PUF60).²² Similarly to U2AF65, PUF60 and the peroxisome proliferator-activated receptor gamma (PPAR-γ) co-activator-1 (PGC-1), CAPER displays a transactivation domain, a region rich in serine–arginine pairs (SR domain) as well as an RNA recognition motif (RRM domain).^19,22 The RRM domain, which consists of 2 highly conserved peptide motifs (ribonucleoprotein [RNP]-1 and RNP-2), confers both RNA and single-stranded cDNA binding activity.¹⁹

Jung et al. (2002) first demonstrated that CAPER directly interacted with ERα/β and c-Jun and potently stimulated their transcriptional activity in vitro.¹⁹ Dowhan et al. (2005) later reported that CAPER also interacted with the progesterone receptor (PR) and stimulated hormone-mediated PR transcriptional activity in vitro.²² Importantly, CAPER appears rather selective for c-Jun, ERα/β, and PR and does not interact with other transcription factors such as p53, the AP-1 component c-fos, the NFκB component p50, serum response factor, retinoic acid receptor (RAR)-α, thyroid hormone receptor (TR)-α/β, PPAR-α, liver X receptor-α/β, farnesoid X receptor, and glucocorticoid receptor, among others.^19,22,23

Interestingly, CAPER was recently shown to be involved in human breast cancer progression.²⁰ Indeed, using a tissue microarray and automated quantivative bioimaging analysis (AQUA), we reported that CAPER protein levels were upregulated during the transition from pre-malignancy to IDC.²⁰ Our immunohistochemical analyses further demonstrated that CAPER was undetectable or expressed at relatively low levels in normal breast tissue and assumed a cytoplasmic distribution.²⁰ In contrast, CAPER was expressed at higher levels in DCIS and IDC specimens, where it assumed a predominantly nuclear distribution.²⁰ However, the functional role of CAPER in human breast cancer initiation and progression remained unknown.

Our present results demonstrate that lentiviral-mediated knockdown of CAPER expression markedly reduced the proliferation of the ER-positive human breast cancer cell line MCF-7 in vitro. Importantly, our results further revealed that knockdown of CAPER expression significantly reduced the tumorigenicity of MCF-7 cells in vivo. Indeed, nude mice injected with MCF-7 cells stably expressing CAPER shRNAs developed smaller tumors than mice injected with MCF-7 cells harboring control shRNAs. This is somewhat in contrast with a recent study by Huang et al. (2011), which reported that overexpression of CAPER in TC-71 Ewing sarcoma cells decreased tumor vessel density and growth in vivo.²⁴ However, the effect of CAPER on TC-71 cell growth was shown to be secondary to alterations in VEGF expression and to a lower VEGF₁₆₅/VEGF₁₈₉ ratio.²⁴ Importantly, while overexpression of CAPER was shown to mediate VEGF alternative splicing and to control the shift from VEGF₁₈₉ to VEGF₁₆₅ in TC-71 Ewing sarcoma cells, it had no significant effect on the proliferation of TC-71 cells in vitro.²⁴ Accordingly, knockdown of CAPER expression was also previously shown to affect the alternative splicing of the VEGF gene in T47D breast cancer cells.²² In a separate study, CAPER has also been shown to interact with the C-terminal transactivation domain (TAD) of the NF-κB oncoprotein v-Rel and to suppress its transforming activity in vitro.²³ However, the region of CAPER involved in the co-activation of v-Rel was shown to be different than the one involved in the co-activation of ERα/β and c-Jun.²³ In fact, v-Rel was reported to interact with the N-terminal domain of CAPER, while ERα/β and c-Jun interact with its C-terminal domain.²³ Interestingly, CAPER expression was also recently shown to be upregulated in human colorectal adenomas and carcinomas.²⁵ The expression of CAPER was reported to affect the survival of colorectal cancer cells.²⁶ Indeed, knockdown of CAPER expression markedly reduced the viability of human colorectal cancer cell lines under both normal culture conditions and upon 5-Fluorodeoxyuridine (5FU)-induced cytotoxicity, suggesting a potential role of CAPER in apoptosis and/or cell senescence.²⁶ Whether increased apoptosis and/or cell senescence contributed to the reduced tumorigenicity of MCF-7 cells harboring CAPER shRNAs observed herein remains to be determined. Therefore, future studies will be necessary to address the possible role of CAPER in modulating breast cancer cell apoptosis and senescence.

The proliferative effects of estrogens on mammary epithelial cells are well known to be mediated, at least in part, through the regulation of genes involved in cell cycle progression.²⁷ Interestingly, both our western blot and immunohistochemical analyses show reduced expression of the cell cycle regulators PCNA, MCM7, and cyclin D1 in tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs. Cyclin D1 expression is one of the rate-limiting steps involved in cell cycle progression.^28,29 In fact, ectopic expression of cyclin D1 in human breast cancer cells is sufficient for the G₁–S phase progression.²⁹ Conversely, inhibition of cyclin D1 function has been shown to arrest cell cycle progression in MCF-7 cells.³⁰ Importantly, cyclin D1 has previously been involved in the mitogenic effects of estrogens. For instance, cyclin D1 expression was shown to be upregulated in serum-starved MCF-7 cells stimulated with mitogenic doses of estrogen.³¹ Conversely, treatment of MCF-7 cells with ER antagonists decreased cyclin D1 expression by 50%.³² Moreover, Cyclin D1 levels were positively correlated with ER levels in primary human breast cancers specimens.³³ Interestingly, in patients with ER-positive breast tumors, high levels of cyclin D1 mRNA were associated with increased risk of relapse, local recurrence, metastasis, and death.³⁴ PCNA and MCM7, 2 well-characterized targets of the retinoblastoma (RB) protein, have also been shown to be implicated in estrogens proliferative effects. PCNA is a DNA polymerase co-factor that is essential for DNA synthesis during the S phase of the cell cycle.^35,36 Treatment of MCF-7 cells with estrogen was shown to increase PCNA levels during the S phase.^31,37 High PCNA expression also predicts early relapse and shorter survival in breast cancer patients.³⁸ On the other hand, MCM proteins are recruited to the cell cycle pre-replication complex and are essential for initiation of DNA synthesis.^39,40 DNA-microarray studies reported increased expression of the MCM7 gene in human breast cancer cells stimulated with estrogen.³¹

Our results also demonstrate reduced levels of the protein synthesis marker phospho-4EBP1 in tumors collected from mice injected with MCF-7 cells harboring CAPER shRNAs. 4EBP1 is the downstream effector of the mammalian target of rapamycin (mTOR) pathway, which plays a critical role in the regulation of ribosomal protein synthesis.^41,42 mTOR-dependent phosphorylation of 4EBP1 allows the release of the eukaryotic initiation factor 4E (eIF4E) and activation of protein synthesis.^41,42 Importantly, treatment of MCF-7 cells with mitogenic doses of estrogen was previously shown to increase the phosphorylation levels of 4EBP1 in vitro.⁴³ The levels of phospho-4EBP1 were also shown to positively correlate with histologic grade, tumor size, lymph node metastasis, and recurrence in breast cancer patients.⁴⁴

Given that CAPER functions as a co-activator of both ERα/β and AP-1/c-Jun,¹⁹ it may confer both estrogen-dependent and estrogen-independent effects. Therefore, although our results suggest that knockdown of CAPER reduces ER transcriptional activity, its role as an AP-1/c-Jun co-activator gene should not be overlooked. AP-1 is a heterodimeric transcription factor composed of several proteins, such as Jun, Fos, and ATF, that are essential for its dimerization and DNA binding.⁴⁵ The AP-1 complex binds the consensus DNA sequence “TGAGTCA” found in a variety of promoters.⁴⁵ AP-1/c-jun is a key regulator of the different processes involved in tumorigenesis, including proliferation, migration, invasion, and metastasis.⁴⁶ Importantly, AP-1/c-Jun has previously been shown to regulate human breast cancer cell proliferation.^47-49 For instance, overexpression of c-Jun was shown to increase the tumorigenic and invasive potentials of MCF-7 cells in both in vitro and in vivo settings.⁴⁷ Conversely, inhibition of AP-1/c-Jun transcriptional activity in MCF-7 cells through expression of a c-Jun dominant-negative mutant (Tam67) was shown to suppress cyclin D1 expression, block cell cycle progression, and inhibit MCF-7 cell proliferation in vitro.^48,49 Inhibition of AP-1/c-Jun transcriptional activity was also shown to suppress MCF-7 cell growth in xenograft mouse models.⁴⁸ Phosphorylation of c-Jun at ser63 and ser73 residues by kinases such as the c-Jun N-terminal kinase (JNK) was shown to be required for AP-1/c-Jun transcriptional activity.^50,51 In addition, interaction of c-Jun with the Jun activation domain-binding protein-1 (Jab1) co-activator was shown to promote c-Jun phosphorylation and enhance AP-1 transcriptional activity.^52-54 Interestingly, our immunoblot and immunohistochemical analyses show reduced levels of both phospho(ser63)-c-Jun and phospho(ser73)-c-Jun in tumors harvested from mice injected with MCF-7 cells stably expressing CAPER shRNAs. Thus, our results suggest that besides its effect on ER signaling, knockdown of CAPER expression might also suppress AP-1/c-Jun transcriptional activity in MCF-7 cells. However, future studies are warranted to determine if CAPER expression could affect the proliferation and/or tumorigenicity of ER-negative human breast cancer cells.

Conclusions

Knockdown of CAPER expression markedly reduced the proliferation of ER-positive human breast cancer cells in vitro. Knockdown of CAPER further decreased mammary tumor growth in orthotopic xenograft mouse models. Mechanistically, knockdown of CAPER abrogated the activity of proliferative and protein synthesis pathways.

Materials and Methods

Materials

MCF-7 cells were purchased from American Type Culture Collection (ATCC, cat# HTB-22). Transduction-ready control shRNA and human CAPER shRNA lentiviral particles were purchased from Sigma-Aldrich (cat# SHC001V and SHCLNV-NM-004902, respectively). A rabbit polyclonal antibody (pAb) to CAPER was purchased from BioVision Inc (cat# 3733-100). Mouse monoclonal antibodies (mAbs) to PCNA (cat# sc-25280), MCM7 (cat# sc-9966), and Cyclin D1 (cat# sc-20044) were purchased from Santa Cruz Biotechnology. A rabbit pAb to phospho(ser65)-4EBP1 (cat# 9451) and rabbit mAbs to phospho(ser63)-c-Jun (cat# 2361) and phospho(ser73)-c-Jun (cat# 3270) were purchased from Cell Signaling Technology. A rabbit mAb to cyclin D1 (cat# ab-16663) was purchased from Abcam. A mouse mAb to glyceraldehyde 3-phosphate dehydrogenase (GAPDH, cat# 10R-G109a) was purchased from Fitzgerald Industries. Rabbit and mouse horseradish peroxidase (HRP)-conjugated secondary antibodies were purchased from BD-Biosciences.

Lentiviral-mediated knockdown of CAPER

MCF-7 cells obtained from ATCC were cultured in Dulbecco minimum essential medium (DMEM, Gibco/Life Technologies, cat# 11995) supplemented with 10% fetal bovine serum (FBS, Gibco/Life Technologies, cat# 16140). After 48 h of culture, normal culture medium was replaced with polybrene (5 μg/ml, Santa Cruz Biotechnology, cat# sc-134220)-containing DMEM medium to increase the efficiency of infection. MCF-7 cells were then infected with 50 μL of transduction-ready control shRNA or human CAPER shRNA lentiviral particles in 5 mL of media for 24 h. MCF-7 cells stably expressing the shRNAs were then selected with puromycin dihydrochloride (2.5 μg/ml, Santa Cruz Biotechnology, cat# 108071).

Cell count

MCF-7 cells stably expressing either control or CAPER shRNAs were seeded at a density of 15 000 cells per well in 12-well culture dishes (Fisher Scientific, cat# 07–200). Seven days later, cells were trypsinized and counted, and the number of cells expressed as a percentage of the control group.

Animals and surgical procedures

This study was conducted according to the guidelines of the National Institutes of Health (NIH) and the Thomas Jefferson University’s Institutional Animal Care and Use Committee (IACUC). Eight-week-old athymic NCR-nu/nu female nude mice were obtained from the National Cancer Institute (NCI) Mouse Repository (cat# 01B74). Female nude mice were anesthetized with 2–5% isoflurane by inhalation. A ventral incision was then performed to expose the fourth inguinal right mammary gland. MCF-7 cells (5 × 10⁶) stably expressing either control or human CAPER shRNAs were resuspended in 50 μL of DMEM supplemented with 10% FBS and injected into the right mammary fat pad with a sterile syringe with a 26-gauge needle (n = 14–20 for each group). The incision was subsequently closed with a 5–0 silk suture (Harvard Apparatus, cat# 510479). Since MCF-7 cells are ER-positive and estrogen-dependent, 17β-estradiol pellets (0.72 mg/pellet; 60-d release; Innovative Research of America, cat# SE-121) were also implanted in female nude mice subjected to mammary fat pad injections. The pellet implantation was performed while the mice were still under anesthesia for the fat pad injections (see above). The site of incision was scrubbed. The skin was lifted on the lateral side of the neck of the animal, and an incision equal in diameter to that of the pellet was performed. Then, a horizontal pocket of about 2 cm beyond the incision site was created, and the pellet was introduced with forceps.

Six weeks after fat pad injections, mice were euthanized by CO₂ inhalation, and tumors were harvested, weighed and measured. Volumes were calculated by assuming that a tumor is an ellipsoid using the formula V = (4/3)*π*a*b², where “V” is the volume, “a” is the length of the long axis, and “b” is the length of the short axis. The tumors collected were cut in half, and each half was either frozen in liquid nitrogen or fixed in 4% PFA for western blotting and immunohistochemical analyses, respectively.

Immunoblot analyses

Cells (n = 4 for each group) and tumors (n = 14 to 20 for each group) were homogenized in a RIPA lysis buffer containing protease and phosphatase inhibitors. The lysates were centrifuged at 12 000 × g for 10 min in order to remove the insoluble debris. The bicinchoninic acid (BCA, Fisher Scientific, cat# PI-23250) reagent was subsequently used to determine the protein concentration of each sample, as well as the volume required for 50 μg of proteins. Proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes. The membranes were placed in blocking solution for 30 min and subsequently washed with 10 mM Tris, 150 mM NaCl, and 0.05% Tween 20 (1X-TBS-Tween). The membranes were incubated with a given primary antibody for 3 h. Finally, horseradish peroxidase (HRP)-conjugated secondary antibodies were used to detect bound primary antibody using the SuperSignal chemiluminescence substrate (Fisher Scientific, cat# 34078). Western blots were quantitated using the NIH ImageJ software (using the mean gray value of each band).

Immunohistochemical analyses

Tumors fixed in PFA (n = 14 to 20 for each group) were embedded in paraffin. Paraffin-embedded 5-μm-thick sections were dehydrated in a series of graded ethanol and incubated in xylene for 15 min. Slides were then gradually rehydrated in ethanol and completely rehydrated in distilled water for 10 min. Antigen retrieval was performed with 10 mM sodium citrate buffer (pH 6.0) for 10 min. Then, sections were incubated with 3% hydrogen peroxide (H₂O₂) for 10 min, washed, and blocked with 10% goat serum in PBS for 1 h at room temperature (RT). After blocking, slides were incubated with a given primary antibody overnight at 4 °C. Then, slides were washed and incubated with biotinylated secondary antibodies (Vector Labs, cat# BA1000 and BA9200) for 30 min at RT. After washing in PBS, slides were incubated with HRP-conjugated streptavidin (Dako, cat# P0397) for 30 min at RT. Slides were then washed and incubated with the 3,3-diaminobenzidine (DAB) reagent (Dako, cat# K3468) until they developed a brown color. Finally, slides were washed, counterstained with Mayer hematoxylin (Sigma, cat# MHS16), dehydrated, and mounted with Permount (Fisher Scientific, cat# SP15-100).

Statistical analysis

All data were expressed as mean ± SEM. Differences between groups were evaluated by either unpaired Student t test or one-way ANOVA followed by Tukey multiple-group comparisons test, where appropriate. Statistical significance was assumed at P < 0.05.

Glossary

Abbreviations:

4EBP1

4E-binding protein-1

AP-1

activator protein-1

AQUA

automated quantivative bioimaging analysis

ASC-2

activating signal co-integrator-2

BCA

bicinchoninic acid

CAPER

co-activator of AP-1 and ERs

CO₂

carbon dioxide

DAB

3,3-diaminobenzidine

DCIS

ductal carcinoma in situ

DMEM

Dulbecco minimum essential medium

eIF4E

eukaryotic initiation factor 4

ER

estrogen receptor

ERE

estrogen response element

FBS

fetal bovine serum

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

GST

glutathione-S-transferase

H₂O₂

hydrogen peroxide

HCC1.4

hepatocellular carcinoma-1.4

HRP

horseradish peroxidase

IACUC

institutional animal care and use committee

IDC

invasive ductal carcinoma

JAB1

Jun activation domain-binding protein-1

JNK

c-Jun N-terminal kinase

mAb

monoclonal antibody

MCM7

mini-chromosome maintenance protein-7

mTOR

mammalian target of rapamycin

pAb

polyclonal antibody

PCNA

proliferating cell nuclear antigen

PFA

paraformaldehyde

PGC-1

peroxisome proliferator-activated receptor gamma (PPAR-γ) co-activator-1

PR

progesterone receptor

PUF60

poly(U)-binding-splicing factor-60

RAR

retinoic acid receptor

RB

retinoblastoma protein

RNP

ribonucleoprotein

RRM

RNA recognition motif

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

SR

serine-arginine

TAD

transactivation domain

TR

thyroid hormone receptor

U2AF65

U2 auxiliary factor