Therapeutic Silencing of Bcl-2 by Systemically Administered siRNA Nanotherapeutics Inhibits Tumor Growth by Autophagy and Apoptosis and Enhances the Efficacy of Chemotherapy in Orthotopic Xenograft Models of ER (−) and ER (+) Breast Cancer

Introduction

The Bcl-2 oncogene is overexpressed in 50–70% of all human cancers, including breast cancers, and is associated with an aggressive clinical course and poor survival.^{1,2,3,4,5,6,7} The Bcl-2 family comprises prosurvival antiapoptotic proteins (Bcl-2, Bcl-x_L, Mcl-1, Bcl-w, and A-1) and proapoptotic proteins (Bax, Bak, Bik, Bad, Bid, HRK, BMF, NOXA, and PUMA).^1,2 The Bcl-2 family can be defined by the presence of conserved motifs known as Bcl-2 homology domains (BH1 to BH4). Bcl-2 contains all four BH domains, whereas the other prosurvival members contain at least BH1 and BH2.¹ The Bcl-2 gene codes for a 25-kDa antiapoptotic protein that promotes cell survival and neoplastic cell expansion.^3,4,5,6,7,8 Inhibition of Bcl-2 enhances the sensitivity of cancer cells to standard therapies,^8,9 thereby indicating the importance of this gene as a potential therapeutic target in various human cancers.

RNA interference, a recently discovered natural process of gene silencing, emerged as an important tool for sequence-specific gene knockdown and is considered to hold great promise for developing targeted molecular therapies for cancer and other diseases associated with increased gene expression as well as viral infections.¹⁰ RNA interference mediated by small interfering RNA (siRNA) can specifically knock down target gene expression via DICER and the RNA-induced silencing complex, causing degradation of the mRNA and preventing the corresponding protein expression.¹⁰ Although siRNA has been shown to target and silence genes, in vivo delivery of siRNA to tumors remains as a great challenge. The major limitations to translate siRNA-based therapies into the clinic include the degradation of siRNA by nucleases after systemic administration, poor cellular uptake, and a lack of effective systemic delivery methods that are safe and nontoxic.¹¹ Although nanocarriers, including liposomes, are not specifically targeted to tumor cells, they passively accumulate in tumor tissues owing to an enhanced permeability and retention effect and the leaky and disorganized nature of angiogenic tumor vasculature. Typically tumor-associated endothelium contains openings varying in size from 50 to >500 nm, in contrast to normal endothelium with pores (fenestra) of 10–50 nm.¹² Cationic (positively charged) liposomes have traditionally been used in in vitro and in vivo gene transfections. However, applications of cationic liposomes in humans are limited because of their toxicity to mammalian cells owing to reactive oxygen species induction and lung inflammation.¹³ If successfully and safely administered, siRNA-based therapies have advantages in drug development over small molecules, biological agents, antisense oligonucleotides and antibodies because they can target “undruggable” targets, which comprise more than two-thirds of the oncogenic targets. Furthermore, siRNA is highly specific, easily synthesized, and cost effective.^11,12 In addition, siRNA-mediated target gene silencing is significantly more potent (more than 100-fold difference in the half maximal inhibitory concentration) and efficient than antisense oligonucleotides or ribozymes.¹⁴

Autophagy is a lysosomal degradation pathway that is a major cellular process for degradation of cytoplasmic organelles and long-lived, misfolded, or damaged proteins.¹⁵ Autophagy is mediated by a set of conserved genes called ATG, including Beclin 1 (ATG6), ATG5 and ATG8 (LC3), and others.¹⁵ Autophagy is induced by nutrient and energy deprivation and metabolic stress and may function as a protective and prosurvival mechanism.¹⁶ Autophagy induction can lead to cell death, also known as autophagic cell death (type II programmed cell death), depending on the cellular context and stimulus.^{15,16,17,18,19,20} Bcl-2 inhibits the autophagic process by physically binding to Beclin-1, an autophagy-promoting protein, and limiting its function.²¹ Inhibition of Bcl-2 leads to autophagic cell death in MCF7 breast cancer cells.¹⁷ Furthermore, recent data suggest that the oncogenic effect of Bcl-2 arises from its ability to inhibit autophagy but not apoptosis, thereby indicating that modulating autophagy may be important in designing anticancer therapies.²²

In this study, we sought to determine whether therapeutic silencing of Bcl-2 by systemic i.v. administration of nanoliposomal siRNA provides effective gene silencing, inhibits tumor growth and further enhances the efficacy of the most commonly used chemotherapeutic agents (doxorubicin and paclitaxel) in both estrogen receptor-negative (ER (−)) and ER-positive (+) orthotopic breast tumors in nude mice. To our knowledge, our findings are the first evidence that in vivo targeting of Bcl-2 suppresses the growth of ER(−) and ER(+) breast tumors in orthotopic xenografts via the induction of both apoptotic and autophagic cell death, thereby suggesting that in vivo inhibition of Bcl-2 is a viable clinically therapeutic approach and may prevent disease progression.

Results

Discussion

In this study, we demonstrated for the first time that in vivo therapeutic targeting of Bcl-2 by i.v. nanoliposomal Bcl-2-siRNA significantly inhibits tumor growth in preclinical models of both ER(−) and ER(+) breast cancers. We also provide the first evidence that therapeutic targeting of Bcl-2 induces autophagy and apoptosis in both ER(−) and ER(+) breast tumors in vivo. Furthermore, silencing of Bcl-2 also significantly increased the efficacy of chemotherapy in both models in vivo.

Bcl-2 is one of the most important and common mediators of survival and drug resistance in most human cancers.^1,2,3,30 Bcl-2 expression leads to aggressive disease course poor survival in patients with different cancers.⁷ Therefore, Bcl-2 is considered an excellent molecular target for therapies for breast and other cancers. However, therapeutic silencing of Bcl-2 in tumors remains a great challenge. Although siRNA-based gene silencing has great potential for molecularly targeted therapies, clinical applications of siRNA-based therapies are hampered by challenges to systemic administration and delivery into tumors.^31,32 When injected systemically, siRNA is rapidly degraded by nucleases in serum and body fluids and cleared from plasma with a half-life of minutes. Therefore, the development of safe and effective in vivo systemic delivery systems for successful clinical applications of siRNA-based therapies is critical.^10,33,34 To therapeutically silence Bcl-2 in breast tumors in vivo, we used liposomes incorporating Bcl-2-specific siRNA that led to significant and robust target gene knockdown in tumors (Figure 2a). A single injection of a small dose of liposomal siRNA (0.15 mg/kg) provided a potent (~80–90%) inhibition of Bcl-2. It is also important to note that the siRNA doses used in our study were about 60- to 120-fold less compared with other reports that used 10 mg/kg siRNA in cationic liposomes,³⁵ and Bcl-2 siRNA was well tolerated in mice. The neutral lipid-based delivery system was safe and effective and produced no obvious toxic effects in the animals during treatment in the current and previous studies.³⁶ However, most commonly used cationic liposomes are highly toxic in vitro and in vivo in mice, thereby limiting their clinic applications.^13,37

The other important finding was that NL-Bcl-2 siRNA treatment significantly enhanced the antitumor efficacy of chemotherapy (Doxorubicin), especially in the ER(−) animal model. However, compared with ER(−) model this effect was slightly less pronounced compared with ER(+) model. This could be related the intrinsic balance between pro- and antiapoptotic proteins (e.g., Bcl-2 vs. Bax) as well as the activity of other signaling pathways such as PI3K/Akt and Ras/Raf/Erk in the ER(−) and ER(+) cancer cells. Although ER(−) cells tend to express less Bcl-2, p53, and K-Ras are mutated in MDA-MB-231 cells compared with ER(+) MCF7 cells.

Autophagy is one of the novel mechanisms of cell death.¹⁶,³⁸,³⁹ Autophagy may function as a survival pathway during nutrient deprivation or starvation.^15,16,19 More importantly, reduced or defective autophagy in mammary tumors activates DNA damage response and synergizes with defective apoptosis to accelerate tumorigenesis.³⁴ We previously showed that inhibition of Bcl-2 induces autophagic cell death in ER(+) MCF-7 breast cancer cells in vitro.¹⁷ The findings of the present study demonstrated that Bcl-2 silencing in ER(+) and ER(−) breast tumors induces autophagy and apoptosis, leading to the suppression of tumor growth (Figure 8). The induction of autophagy by doxorubicin was also mediated by Bcl-2 downmodulation, leading to Beclin-1, ATG5 and LC3-II induction (Figure 8). More importantly, Bcl-2 siRNA contributes to cell death, as knockdown of authophagy genes prevented the induction of cell death and increased cell survival (Figures 6a, 8).

The induction of autophagy following Bcl-2 silencing may be mediated by two different mechanisms in breast cancer cells: (i) inhibition of Bcl-2 relieves its suppressor activity on Beclin-1, which is physically bound and blocked by Bcl-2²¹ and (ii) the higher apoptotic threshold owing to the lack of caspase 3 and p53 mutations in MCF-7 and MDA-MB231 cells, respectively, may favor the induction of autophagy as a default cell death mechanism.⁴⁰ In fact, studies of Bax/Bak double-knockout MEF cells showed that cells undergo autophagy-mediated cell death in response to a death stimulus, suggesting that autophagy may function as an alternative cell death mechanism when the apoptotic threshold is increased or blocked; thus, this type of cell death may be used in eliminating cancer cells that are resistant to apoptosis.^6,41 More important, a recent study Oh et al. demonstrated that the antiautophagic property of Bcl-2 is the key feature of Bcl-2-induced tumorigenesis.²² They reported that the growth-promoting activity of Bcl-2 is strongly correlated with its suppression of autophagy in xenograft tumors. MCF-7 cells expressing mutated Bcl-2 that was defective in inhibiting apoptosis but competent for suppressing autophagy grew in vitro and in vivo as efficiently as wild-type Bcl-2-expressing cells, indicating that the oncogenic effect of Bcl-2 arises from its ability to inhibit autophagy but not apoptosis.²²

Tumors derived from cells that overexpress Bcl-2 grow more aggressively in vivo. This could be attributed to events other than the antiapoptotic and antiautophagic properties of Bcl-2. In fact, emerging studies suggest that Bcl-2 promotes cancer progression by enhancing cell invasion, cell migration, and the metastatic potential of various cancer types.^27,28,29 We observed that Bcl-2 downregulation reduced the activity (phosphorylation) of FAK/SRC, HIF-1α, and cyclin D1 in tumor xenografts (Figure 7). FAK is known to play a major role in cell migration, invasion/metastasis, and drug resistance by activating the Ras/MEK/ERK5 and PI3K/Akt survival pathways.^42,43,44 Future studies should investigate in detail how Bcl-2 regulates cell migration, invasion, and angiogenesis and cell cycle in breast tumors in vivo. HIF-1α is a mediator of cellular response to hypoxia and is associated with increased angiogenesis, metastasis, treatment resistance, and poor prognosis.²⁰ Anai et al. recently showed that inhibition of Bcl-2 leads to reduced angiogenesis in human prostate tumor xenografts.²⁴ In addition, Bcl-2 overexpression increases vascular endothelial growth factor promoter activity through the HIF-1α transcription factor,²⁵ thereby providing a link between Bcl-2 and angiogenesis.^20,26 Breast cancer patients with a higher Ki-67 have been shown to have significantly poorer prognosis, early recurrence, and reduced overall survival rates.⁴⁵ Inhibition of Ki-67 expression in tumors after Bcl-2 siRNA treatment suggests that overall treatment response and antitumor effects may be due to multiple mechanisms, including apoptosis and autophagy.

Pretreatment with Bcl-2 antisense enhanced the antitumor activity of various chemotherapeutic agents, such as cyclophosphamide, dacarbazine, and docetaxel, in several cancers in vitro.⁴⁶ George et al. reported that in vitro treatment of human glioma cells with Bcl-2 siRNA and taxol (100 nmol/l) increased the apoptotic cells in a TUNEL assay up to 70% compared with 30% in those treated with taxol alone (100 nmol/l).⁴⁷ Our in vitro and in vivo findings suggest that targeting Bcl-2 is a highly effective therapeutic strategy for enhancing the efficacy of standard chemotherapeutic agents in breast cancer.

In conclusion, our study suggests that highly specific targeting of Bcl-2 by siRNA-based therapies provides efficient and potent target silencing, inhibits tumor growth, and enhances the efficacy of standard chemotherapies. Thus, targeting Bcl-2 by systemically administered siRNA is a viable approach and may be beneficial to breast cancer patients with Bcl-2-overexpressing tumors.

Materials and methods

Cell lines, culture conditions, and reagents. ER(−) MDA-MD-231 and ER (+) MCF-7 human breast cancer cell lines were purchased from ATCC. Doxorubicin-resistant MCF7 cells were obtained from Dr. Kapil Mehta (The University of Texas MD Anderson Cancer Center, Houston, Texas, USA). MDA-MB-231/luciferase cells were transduced with luciferase-expressing lentiviral particles that were prepared in our laboratory.²³ Transfection of MDA-MB-231 cells was performed at a multiplicity of infection virus particle concentration of 5. Five days after transfection, the cells were treated with 5 µg of puromycin concentration to select stably luciferase expressing cells. All cell lines were cultured at 37 °C in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum in a humid incubator with 5% CO₂. For the cell proliferation experiments, cells were seeded at a density of 1–2 × 10⁵ cells in T-25 tissue culture flasks or 1.25–3 × 10³ in 96-well plates. Cells were collected via trypsin exposure, and cell numbers were determined using a Neubauer cell-counting chamber. All experiments were repeated at least three-times.

Cell viability and growth assays. Cell viability in response to siRNA treatments or doxorubicin was evaluated using a tetrazolium/formazan (MTS)-based CellTiter 96 AQ_ueous One Solution cell proliferation assay (Promega, Madison, WI).¹⁷ Cells were seeded in 96-well plates at a density of 3 × 10³ cells per well in 100 µl of medium. The next day, the medium was removed, and cells were transfected with siRNA (50 nmol/l) in 100 µl of medium plus transfection mix or treated with doxorubicin for 72 hours. Plates were read at wavelength of 490 nm in a VMax kinetic enzyme-linked immunosorbent assay microplate reader (Molecular Devices Corporation, Sunnyvale, CA, USA). The dead and viable cells were also detected via a trypan blue exclusion assay in which viable cells are able to exclude the dye and remain unstained while dead cells take up the blue coloring agent.

Clonogenic assay. This assay is an in vitro cell survival and proliferation assay based on the ability of a single cell to grow into a colony.^18,36 Briefly, 500 cells were mixed gently and plated on a 6-well plate. After being incubated for 24 hours, the cells were transfected with control and Bcl-2 siRNA every 5 days, and about 2 weeks later, the cells were washed with phosphate-buffered saline and stained with crystal violet. Colonies with a diameter of more than 50 cells were counted. The experiment was repeated three-times.

siRNA transfections. Exponentially growing untreated MCF-7 and MDA-MB-231 cells were collected and plated (2 and 1.5 × 10⁵/flask in 4 ml, respectively) 24 hours before transfection. Plated cells were transfected with either Bcl-2 siRNA or control siRNA (50 nmol/l). siRNA sequences targeting Bcl-2 and nonsilencing control siRNA 5′-AACATCGCCCTGTGGATGACT-3′ and 5′-AATTCTCCGAACGTGTCACGT-3′, respectively.¹⁷ Beclin-1 siRNA and ATG8 siRNA²² were used. The siRNAs were dissolved in sterile buffer provided by the manufacturer (all from Qiagen Inc, Valencia, CA, USA). On the day of transfection, 1.5 µg of siRNA was mixed with HiPerFect transfection reagent according to the manufacturer's instructions (Qiagen) and added to the cells in each well.

Western blot analysis. After treatment, the cells were trypsinized and collected by centrifugation, and whole-cell lysates were obtained using a lysis buffer as described previously.⁴⁸ Total protein concentration was determined using a protein assay kit (Bio-Rad, Hercules, CA, USA). Aliquots containing 30 µg of total protein from each sample were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis with a 4–20% gradient gel and transferred to polyvinylidene difluoride membranes. The membranes were blocked with 5% dry milk in Tris-buffered saline with Tween 20 (TBST) and probed with primary antibodies of human specific Bcl-2 monoclonal antibody (Dako Cytomation California Inc., Carpinteria, CA, USA), Beclin-1, ATG5 (Santa Cruz, CA, USA), LC3 (Axorra LLC, San Diego, CA, USA), human specific monoclonal cleaved poly(ADP-ribose polymerase (PARP; #9546), and cleaved caspase 9 (#9590, Cell Signaling Technology, Beverly, MA, USA). The antibodies were diluted in TBST containing 2.5% dry milk and incubated at 4 °C overnight. After the membranes were washed with TBST, they were incubated with horseradish peroxidase-conjugated antirabbit or antimouse secondary antibody (Amersham Life Science, Cleveland, OH, USA). Mouse anti-β-actin and donkey antimouse secondary antibodies (Sigma–Aldrich, St. Louis, MO, USA) were used to monitorβ -actin expression to ensure equal loading of proteins. Chemiluminescent detection was performed with ChemiGlow detection reagents (Alpha Innotech, San Leandro, CA, USA). The blots were visualized with a FluorChem 8900 imager and quantified with a densitometer using an AlphaImager system (Alpha Innotech).

In vivo detection of apoptosis via TUNEL assay. Apoptotic cells in tumor tissue were detected by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining using an apoptotic cell detection kit following the manufacturer's directions (Promega, Madison, WI, USA).³⁶ Images of the sections were captured by a microscope (Nikon, Tokyo, Japan). The apoptotic index was calculated by dividing the number of TUNEL-positive cells by the total number of cells in the field. Light microscopy was used to count the number of TUNEL-positive cells on ten randomly selected fields for each section.

Evaluation of autophagy via detection of acidic vesicular organelles. Cells were stained with acridine orange as described previously¹⁸ to detect and quantify acidic vesicular organelles. The number of acridine orange-positive cells was determined via fluorescence-activated cell sorting (FACS) analysis. Cell morphology was examined using a phase-contrast microscope (Nikon, Melville, NY, USA) while the cells remaining in their culture flasks.

Nanoliposomal siRNA preparation. Control siRNA and Bcl-2 siRNA were encapsulated using 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine-lipid–based nanoliposomal particles. Briefly, siRNA was mixed with the lipid at a ratio of 1:10 (w/w). Tween 20 was added to the mixture at a ratio of 1:19 Tween 20: siRNA/lipid in the presence of excess tertiary butanol.³⁶ After being vortexed, the mixture was frozen in an acetone/dry ice bath and lyophilized. Before animals were injected, the lyophilized lipid-siRNAs were reconstituted with 0.9% saline to form liposomes and sonicated for 3 minutes. The mean size of the liposomes incorporating the siRNAs was measured using a Zetasizer Nano ZS (Malvern, Worcestershire, UK) and found to be about 65 nm with zeta potential of 1.9 ± 0.24 for NL-empty and −2.7 ± 0.33 for NL-cont siRNA in phosphate-buffered saline. Free siRNA was separated from liposomes using filter units with a 30,000 nominal molecular weight limit (Millipore Corp., Billerica, MA, USA). The liposomal suspension was added to the filters and centrifuged at 5,000×g for 40 minutes at room temperature. Fractions were collected, the material trapped in the filter was reconstituted with 0.9% saline, and the siRNA of the collected fraction and the elute were measured via spectrophotometry.

Tumor models in mice. Athymic female nude mice (NCr nu/nu) mice 5–6-weeks old were obtained from the Department of Experimental Radiation Oncology at MD Anderson. The mice were housed three per cage in standard acrylic glass cages in a room maintained at a constant temperature and humidity with a 12-hour light-dark cycle. They were fed a regular autoclaved chow diet with water ad libitum. All studies were conducted according to an experimental protocol approved by the MD Anderson Institutional Animal Care and Use Committee. ER(−) MDA-MB-231 cells (1.5 × 10⁶) and ER(+) MCF7 cells (7.0 × 10⁶) were orthotopically injected into the right mammary fat pat of each mouse. For the experiments using MCF-7 cells, mice were primed with 17β-estradiol applied subcutaneously (1.7 mg estradiol/pellet) under the left shoulder to promote tumor growth. When tumor size reached 3–5 mm about 2 weeks later, mice were administered liposomal siRNA and doxorubicin once a week.

Evaluation of in vivo growth of tumors after systemic liposomal siRNA treatments. MDA-MB-231 and MCF-7 cells were implanted orthotopically in the mammary fat pads of athymic nude mice (NCr nu/nu) that were 5–6-weeks old. Two weeks tumor cell injection, luciferase activity was measured by injecting d-luciferin potassium salt (Molecular Probes, Eugene, OR, USA) using an IVIS imaging system (Xenogen, Alemeda, CA, USA) as previously described.²³ Briefly, the mice were anesthetized, and d-luciferin was injected at 100 mg/kg mouse body weight. Ten minutes after d-luciferin injection, the mice were imaged with an IVIS Imaging System 2000 coupled with data acquisition controlled by a computer running LivingImage software (Xenogen, Alameda, CA, USA).²³ Mice with equally sized tumors were randomly assigned to one out of four treatment groups: group I received nanoliposomal (NL)-control siRNA (0.15 mg siRNA/kg) twice weekly via intravenous (i.v.) injection; group II received NL-Bcl-2-siRNA (0.15 mg siRNA/kg) twice weekly via i.v. injection; group III received both control NL-siRNA (0.15 mg siRNA/kg) and doxorubicin (4 mg/kg) weekly via intraperitoneal (i.p.) injection; and group IV received both NL-Bcl-2-siRNA (0.15 mg siRNA/kg) twice weekly via i.v. injection and doxorubicin (4 mg/kg) weekly via i.p. injection.³⁶ The resulting tumor growth was assessed after 4 weeks (eight doses) of treatment using the IVIS imaging system. The mice were euthanized 48 hours after the final injection, and primary tumors were excised and weighed. A portion of the tumors was in liquid nitrogen for molecular analysis and another portion was formalin fixed and paraffin embedded. In any instance, please clarify how liquid nitrogen was used for immunohistochemistry for routine hematoxylin and eosin staining and TUNEL assay as described previously.³⁶ The remaining tumor tissue was stored at −80 °C until use.

Statistical analysis. The data were expressed as the means ± SD of three or more independent experiments, and statistical analysis was performed using the two-tailed and paired Student's t test. P < 0.05 was considered statistically significant and indicated by an asterisk.

SUPPLEMENTARY MATERIALFigure S1. Dose-dependent downregulation of Bcl-2 protein in MDA-MB231 tumors after single NL-Bcl-2 siRNA injection (iv. tail vein).Figure S2. Therapeutic silencing of Bcl-2 by only three i.v. injections of NL-Bcl-2 siRNA inhibits in vivo tumor growth of ER(-) MDA-MB-231 xenografts in nude mice (*p<0.05).Figure S3. Treatment schedules with siRNA and chemotherapy in mice bearing tumors.Figure S4. A) Dose-dependent inhibition of MDA-MB-231 cells by doxorubicin (72h). B) Doxorubicin induces autophagy in MDA-MB-231 cells as indicated by acridine orange staining and FACS analysis (48h). C) Doxorubicin induces apoptosis and autophagy in MDA-MB-231 cells as indicated by Annexin V/PI and acridine orange staining and FACS analysis (48h). D) Knockdown of autophagy genes including ATG5 and Beclin 1 inhibits doxorubicin-induced autophagy in MDA-MB-231 cells.