• Open Access

Knocking down cyclin D1b inhibits breast cancer cell growth and suppresses tumor development in a breast cancer model

Authors


4To whom correspondence should be addressed. E-mail: qlgu@shsmu.edu.cn

Abstract

Cyclin D1 is aberrantly expressed in many types of cancers, including breast cancer. High levels of cyclin D1b, the truncated isoform of cyclin D1, have been reported to be associated with a poor prognosis for breast cancer patients. In the present study, we used siRNA to target cyclin D1b overexpression and assessed its ability to suppress breast cancer growth in nude mice. Cyclin D1b siRNA effectively inhibited overexpression of cyclin D1b. Depletion of cyclin D1b promoted apoptosis of cyclin D1b-overexpressing cells and blocked their proliferation and transformation phenotypes. Notably, cyclin D1b overexpression is correlated with triple-negative basal-like breast cancers, which lack specific therapeutic targets. Administration of cyclin D1b siRNA inhibited breast tumor growth in nude mice and cyclin D1b siRNA synergistically enhanced the cell killing effects of doxorubicin in cell culture, with this combination significantly suppressing tumor growth in the mouse model. In conclusion, the results indicate that cyclin D1b, which is overexpressed in breast cancer, may serve as a novel and effective therapeutic target. More importantly, the present study clearly demonstrated a very promising therapeutic potential for cyclin D1b siRNA in the treatment of cyclin D1b-overexpressing breast cancers, including the very malignant triple-negative breast cancers. (Cancer Sci 2011; 102: 1537–1544)

Cyclin D1 (CycD1), encoded by cyclin D1/CCND1 and overexpressed in a broad range of solid malignancies, is an important cell cycle regulator that promotes G1/S phase transition by activation of Cdk4 and Cdk6 kinase activity.(1,2) Expression of CycD1 in normal dividing cells is upregulated in the late G1 phase by transcription activation through E2F family transactivators.(3–5) The CycD1 accumulated during the G1/S transition simultaneously forms complexes with Cdk4 and Cdk6 and subsequently promotes initiation of DNA replication and centrosome duplication.(6–8) The abundant CycD1 eventually becomes phosphorylated and destroyed by ubiquitin-mediated proteolysis, which allows normal cell cycle progression. However, the level of CycD1 and the activity of the CycD1–Cdk4 and CycD1–Cdk6 complexes can be regulated aberrantly, with excessive CycD1–Cdk4 and CycD1–Cdk6 activity, in turn, driving cells to replicate their DNA prematurely, resulting in genome instability and tumorigenesis.(9,10) Emerging data from a number of experimental systems indicate that CycD1 also has multiple, kinase-independent cellular functions. First, CycD1 assists in sequestering CDK inhibitors (e.g. p27kip1) and thus bolstering late G1 phase CDK activity.(11) Second, CycD1 is known to bind and modulate the function of several transcription factors that are significant in human cancers.(12) Thus, CycD1 impinges on several distinct pathways that govern cancer cell proliferation. Although intragenic somatic mutation of CycD1 in human disease is rare, CycD1 gene translocation, amplification, and/or overexpression are frequent events in selected tumor types. In addition, a polymorphism in the CycD1 locus that may affect splicing has been implicated in increased cancer risk or poor outcome.(13) Recent functional analyses of an established CycD1 splice variant, namely cyclin D1b (CycD1b), revealed that CycD1b harbors a stronger oncogenic potential than another full-length variant of CycD1, referred to as cyclin D1a.(14,15) However, the molecular mechanisms of CycD1b-driven tumorigenesis have not been fully elucidated. Importantly, total levels of CycD1b in tumor tissues are inversely correlated with survival in patients with breast cancer;(16,17) patients with tumors that have high levels of CycD1b at Stage I die within 5 years of diagnosis, whereas patients with low CycD1b expression in tumors have a much longer survival. These observations indicate that overexpression of CycD1b may be an important cause of breast cancer mortality and that CycD1b may be an important therapeutic target for the development of anticancer drugs.

The RNAi system is used to regulate gene. The functional mediator of RNAi is a 21-nucleotide (nt) siRNA generated by the cleavage of double-stranded (ds) RNA via a complex consisting of Dicer, TAR RNA-binding protein (TRBP), and protein activator of protein kinase PKR (PACT).(18–20) In fact, in recent years, chemically synthesized siRNA oligonucleotides have been demonstrated to be superior agents to conventional antisense oligonucleotide approach that can effectively knock down gene expression by sequence-specific degradation of complementary mRNA in cell culture.(21) Compared with conventional antisense oligonucleotide approaches, siRNA-mediated gene knock down is much more specific and potent. The selection of targeting sequences for siRNA is less restricted, so the rate of producing effective duplexes is higher.(22,23) In addition, siRNA is a double-stranded RNA, which is more resistant to nuclease degradation and it therefore has prolonged stability in in vivo studies.(24,25) These unique properties make siRNAs potential treatments for cancer by targeting mutation- or overexpression-activated oncogenes in cancers. Numerous studies have shown that siRNA can effectively suppress oncogene expression in cancer cells and some siRNA cancer therapies are, indeed, in the preclinical or early stages of clinic trials.(26–28) In the present study, to investigate whether CycD1b can serve as a novel therapeutic target and whether an siRNA-based approach can effectively treat CycD1b-overexpressing breast cancer, we used CycD1b siRNA to target CycD1b overexpression and assessed its ability to suppress breast cancer growth in nude mice. The results clearly demonstrated a very promising therapeutic potential for CycD1b siRNA in the treatment of CycD1b-overexpressing breast cancer, including the highly malignant triple-negative breast cancer.

Materials and Methods

Cell culture.  All cell lines used were obtained from the American Type Culture Collection (Rockville, MD, USA) and were cultured at 37°C in a 5% CO2 incubator. The immortalized normal human mammary epithelial cell line MCF-10A was cultured as described previously.(29) Breast cancer cell lines, including the basal-type cell lines MDA-MB436 and MDA-MB157 (estrogen receptor [ER] and progesterone receptor [PR] negative) and the luminal-type cell lines SK-BR3 (ER and PR negative and overexpressing human epidermal growth factor receptor 2 [HER2 ]), MDA-MB453 (ER and PR negative), and T47D (ER and PR positive), were cultured in DMEM supplemented with 10% FBS, 1%l-glutamine, and 1% penicillin/streptomycin (Sigma, St Louis, MO, USA). For MDA-MB436, 10 mg/mL insulin was also added to the culture medium.

Transfection with siRNA oligonucleotides.  The siRNA oligonucleotides for CycD1b and luciferase (Luc) were synthesized by Qiagen (Valencia, CA, USA). The breast cancer cells (1 × 105/well) in six-well plates were transfected with siRNA oligonucleotides (0.3 μg/well) using Oligofectamine reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions.

Western blotting analysis.  Western blotting assays were performed as what previously described.(17,30) Forty-eight hours after transfection, cells were lysed into mammalian cell lysis buffer (Pierce, Rockford, IL, USA) supplemented with Halt Protease Inhibitor Cocktail (Thermo, MA, USA). After quantified as previously described,(17,30) lysates (50 μg) were analyzed by immunoblotting. The anti-CycD1b polyclonal antibody used in the present study was obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA), whereas the anti-GADPH antibody was from Sigma. The density of each protein band in each sample was determined by densitometry and standardized against GADPH as an internal control. The standardized density of CycD1b in the SK-BR3 cells was arbitrarily set at 100%, with relative levels of CycD1b in other samples calculated by comparing the GADPH-normalized densities against that of the SK-BR3 cells. The data are shown as the mean ± SD of three independent experiments.

Apoptosis and cell cycle analysis.  Standard FACS analysis was used to determine the apoptosis of cells or the phase distribution of cells in the cell cycle, as described previously.(31,32)

Colony formation assay in soft agar.  The standard colony formation assay was performed as described previously.(31) Briefly, breast cancer cells were transfected without (mock) or with siRNA oligonucleotides targeting CycD1b or Luc. Two days after transfection, cells (1 × 103 cells/well) were plated in 24-well plates in culture medium containing 0.35% agar overlying a 0.7% agar bottom layer and cultured at 37°C with 5% CO2. Two to three weeks later, the top layer of the culture was stained with p-iodonitrotetrazolium violet (1 mg/mL) and colonies >100 μm in diameter were counted. To monitor the cell viability of each group, cells (1 × 103 cells/well) were plated in 10-cm culture plates with normal complete medium and, 2–3weeks later, colonies were stained with 1% crystal violet and counted. Finally, the number of colonies in soft agar for the Luc or CycD1b groups was standardized against that for the control group (mock cell; set at 100%).

In vivo assays.  In in vivo experiments, 1 × 106 MDA-MB436, SK-BR3, or T47D cells, mixed with Matrigel Matrix (BD Biosciences, San Jose, CA, USA), were injected into the second pair of mammary glands of nude mice (two injection sites per mouse; 45 mice for each cell line). Two weeks later, nude mice with a tumor burden were randomly divided into three groups (= 15 for each group) and then treated without (mock) or with the an siRNA complex (Luc siRNA or CycD1b siRNA) by intratumoral injection. Each complex contained 10 μg siRNA (for mock injections, mice were injected with PBS instead of siRNA) and 7.5 μL Oligofectamine (Invitrogen) in PBS, which was mixed according to the manufacturer’s instructions. Mice were treated weekly for 4 weeks and were killed either 1 week after the last treatment (for tumor size comparisons) or 2 days after the last treatment (for anti-CycD1b immunohistochemistry or TUNEL assays). Tumor size was measured just before each treatment or after mice had been killed and the tumor volume was determined as described previously.(30)

To assess the effectiveness of combination treatment with CycD1b siRNA and doxorubicin (Dox), MDA-MB436 cells were injected into the second pair of mammary glands of the nude mice as described above (= 70). Beginning on Day 14 after tumor cell injection, mice with a tumor burden were randomly divided into five groups (= 14 per group): mock (liposome alone), Luc siRNA, CycD1b siRNA, Dox (alone), and Dox+CycD1b siRNA. For the mock, Luc siRNA, and CycD1b siRNA groups, mice were treated with repeated weekly intratumoral injections for 4 weeks of liposome alone, Luc siRNA (10 μg), or CycD1b siRNA (10 μg), respectively. Mice in the Dox group were injected weekly with Dox (2 mg/kg, i.p.) for 4 weeks. Mice in the Dox+CycD1b siRNA group were treated for 4 weeks with weekly injections of Dox (2 mg/kg, i.p.), followed the following day with CycD1b siRNA treatment. All these experiments were repeated three times.

All animal protocols performed in the present study were approved by the Institutional Animal Care and Use Committee of Shanghai Jiaotong University. Nude mice, aged 4–5 weeks, were used for the in vivo studies.

Immunohistochemistry.  Tumors were dissected 2 days after the last mock or siRNA treatment and then sectioned, deparaffinized, rehydrated, and stained with anti-CycD1b antibody (Santa Cruz Biotechnology) according to manufacturer’s instructions and then assayed using the Pierce BCA Protein Assay kit (Thermo Scientific, MA, USA). For each treatment group, at least two tumor samples and two slides per sample were analyzed. For each slide examined, 1000 cells were counted from five fields at ×200 magnification and the number of CycD1b-positive cells as a percentage of total cells determined.

TUNEL assay.  Apoptotic cells were identified using the TUNEL labeling kits from Roche (Indianapolis, IN, USA), according to the manufacturer’s instructions. Brown-stained (apoptotic) cells were counted under the microscope at ×200 magnification. The apoptotic index (AI) was defined as the percentage of brown cells among total cells for each sample. For each cell line analyzed, 200 cells were counted; for each tumor sample analyzed, 1000 cells were counted from each fields. Experiments were repeated three times.

Synergistic effects of CycD1b siRNA and Dox treatment of cultured cells.  SK-BR3 and MDA-MB436 cells (1 × 104 cells/well) were transfected with different concentrations of CycD1b siRNA (0, 5 μg, 10 μg, 20 μg, 50 μg, 100 μg/well). Then, 24 h after transfection, cells were split into 96-well culture plates and treated with different concentrations of Dox for 8 days (0, 2 ng, 5 ng, 20 ng, 50 ng, 100 ng/well). The number of viable cells was determined using the MTT assay, as described previously.(33) The synergistic inhibitory effects of CycD1b siRNA and Dox were determined using combination indices of IC80, IC90, and IC95 using the combination index (CI) isobologram method developed by Chou,(34,35) where a CI of 1 indicates an additive effect, a CI > 1 indicates synergism, and a CI < 1 indicates antagonism.

Statistical analysis.  Statistical analyses were performed using two-tailed Student’s t-test. Data are presented as the mean ± SD and differences were considered significant at = 0.05.

Results

CycD1b overexpression is suppressed in breast cancer cells by siRNA targeting.  To determine whether CycD1b can serve as a novel therapeutic target for breast cancer, we first used siRNA oligonucleotides to deplete CycD1b expression in breast cancer cells. The siRNA oligonucleotides were transfected into three CycD1b-overexpressing breast cancer cell lines (SK-BR3, MDA-MB157, and MDA-MB436) and two cell lines in which CycD1b expression is low (T47D and MDA-MB453). Protein levels of the full-length CycD1b (50 kDa) were reduced by up to 96% by CycD1b siRNA in all five cell lines (Fig. 1). The inhibitory effect of the CycD1b siRNA was shown to be specific because a control siRNA oligonucleotide targeting firefly luciferase mRNA (Luc) had no effect on CycD1b expression (Fig. 1). Moreover, siRNA oligonucleotides did not produce non-specific downregulation of gene expression, as demonstrated by the GADPH control (Fig. 1). These data indicate that CycD1b siRNA can effectively suppress the overexpression of CycD1b.

Figure 1.

 Cyclin D1b (CycD1b) siRNA is able to suppress CycD1b overexpression in breast cancer cell lines. Five cell lines (SK-BR3, MDA-MB157, MDA-MB436, MDA-MB453, and T47D) were transfected with siRNA oligonucleotides targeting either CycD1b or luciferase (Luc; control). (a) The protein lysates of the transfected cells were subjected to western blotting analysis using anti-CycD1b and anti-GADPH antibodies (top panel). (b) Relative levels of CycD1b in each cell line was determined after transfection of mock (▪), CycD1b (inline image), or Luc (□) siRNA, as described in the Materials and Methods.

Induction of apoptosis and cell cycle arrest in the G1 phase following siRNA-mediated depletion of CycD1b overexpression.  Because CycD1b and its functional complex CycD1b–Cdk2 play pivotal roles in the G1/S phase transition, we examined changes in the cell cycle distribution and apoptosis in CycD1b-depleted cells. To determine whether depletion of CycD1b promotes tumor cell death, flow cytometry was performed after siRNA transfection. Cells were analyzed at different time points after transfection (72 and 96 h) and significant sub-G1 (apoptotic) populations were observed at 96 h in CycD1b-overexpressing cells (SK-BR3, MDA-MB157, and MDA-MB436). Approximately 14% of these cells underwent apoptosis after transfection of CycD1b siRNA (Fig. 2a). In contrast, only 4% of the same cell lines underwent apoptosis in the mock or Luc siRNA-treated groups (Fig. 2b). The CycD1b-overexpressing cells shrank, became rounder, and detached from the plates 3 days after transfection of CycD1b siRNA, whereas control siRNA-treated cells remained attached on the plates and exhibited normal morphology, also suggesting that apoptosis had occurred. In addition, we confirmed CycD1b siRNA-induced apoptosis in CycD1b-overexpressing cells using the TUNEL assay (Fig. 2c). In contrast with CycD1b-overexpressing cells, we did not observe significant apoptosis in cells in which CycD1b expression was low (MDA-MB453 and T47D) 96 h after transfection (Fig. 2b,c), even though CycD1b protein levels in these cells were suppressed by CycD1b siRNA treatment (Fig. 1). These data indicate that depletion of CycD1b specifically triggers apoptosis in CycD1b-overexpressing cells.

Figure 2.

 Apoptosis and decreased cell proliferation induced by depletion of cyclin D1b (CycD1b) in CycD1b-overexpressing breast cancer cells (SK-BR3, MDA-MB157, and MDA-MB436). (a,b) Downregulation of CycD1b promotes apoptosis of SK-BR3, MDA-MB436 and MDA-MB157 cells (a), but not of T47D and MDA-MB453 cells (b), in which CycD1b is low. Apoptosis was determined by flow cytometry 48 h after transfection of mock (▪), CycD1b (□), or luciferase (Luc [control]; inline image) siRNA. (c,d) Apoptosis was confirmed by the TUNEL assay 48 h after Luc or CycD1b siRNA transfection. (c) Apoptotic cells (brown) were detected in the MDA-MB436 cell line, but not in the T47D cell line. (d) Percentage of apoptotic cells in the MDA-MB436 and T47D cell lines following Luc (□) or CycD1b (▪) siRNA transfection. (e) Cell cycle distribution, as determined by flow cytometry, in the five breast cancer cell lines (= 3 for each group). (□), G1/G0 phase; (inline image), S phase; (▪), G2/M phase. The S phase population was decreased by CycD1b siRNA in CycD1b-overexpressing cells, but not in cells in which CycD1b expression was low.

Furthermore, 48 h after CycD1b siRNA transfection, we observed increased G0/G1 and decreased S phase populations in all three CycD1b-overexpressing cell lines tested, but not in the cells in which CycD1b expression was low (MDA-MB453 or T47D; Fig. 2e). Together, these results indicate that CycD1b siRNA has a specific effect on CycD1b-overexpressing breast cancer cells by promoting apoptosis as well as inhibiting G1/S phase transition.

Depletion of CycD1b inhibits proliferation and transformation in CycD1b-overexpressing breast cancer cells.  To determine whether CycD1b siRNA actually affects proliferation, we examined the growth curves of three CycD1b siRNA-treated groups. Growth was significantly suppressed in the CycD1b-overexpressing cell lines and the two cells lines in which CycD1b expression was low following CycD1b siRNA treatment. As shown in Fig. 3(a), CycD1b siRNA significantly decreased cell numbers in all three CycD1b-overexpressing cell lines (SK-BR3, MDA-MB157, and MDA-MB436) compared with control. However, for the two cells lines in which CycD1b expression was low (T47D and MDA-MB453) or normal breast epithelial cells (MCF10A), cell number was not significantly affected after CycD1b siRNA transfection (Fig. 3a). In addition, CycD1b depletion inhibited the transformation phenotype of CycD1b-overexpressing cells. As shown in Fig. 3(b), CycD1b siRNA significantly reduced the ability of SK-BR3 cells to grow on soft agar, a well known assay that measures cell transformation. This suppression of transformation was also observed for the MDA-MB157 and MDA-MB436 cell lines following depletion of CycD1b expression (data not shown). In contrast, we did not observe significant inhibitory effects on the transformation of T47D (Fig. 3b) or MDA-MB453 cells (data not shown) following CycD1b siRNA targeting. Together, these results show that CycD1b siRNA can significantly inhibit proliferation and transformation of CycD1b-overexpressing breast cancer cells.

Figure 3.

 Inhibition of cell transformation by cyclin D1b (CycD1b) siRNA in vitro. (a) Growth curves for breast cancer (SK-BR3, MDA-MB157, MDA-MB436, MDA-MB453, and T47D) and normal epithelial (MCF-10A) cells following CycD1b siRNA transfection. Data are the mean ± SD (= 3). (b) Suppression of colony formation in soft agar by CycD1b siRNA. SK-BR3 and T47D cells were transfected with mock (inline image), CycD1b (inline image), or luciferase (Luc; inline image) siRNA and then seeded in 0.35% agar containing DMEM and 10% FBS. Cell viability was determined by the colony formation assay, with the number of colonies following Luc or CycD1b siRNA transfection given as a percentage of that following mock siRNA transfection (set at 100%). Data are the mean ± SD (= 3). *= 0.001 compared with mock transfection.

CycD1b siRNA treatment suppresses tumor growth in vivo.  To determine whether CycD1b siRNA could suppress breast cancer growth in vivo, we established breast tumors in nude mice and then treated the mice with CycD1b siRNA. Because MDA-MB436 cells can grow easily in nude mice, we selected this CycD1b-overexpressing cell line to establish the tumor model. As shown in Fig. 4(a), tumor growth in the CycD1b siRNA-treated group was significantly inhibited compared with the control group. To further assess whether CycD1b siRNA was able to suppress the tumor growth, mice were not treated with the siRNA until the tumor burden at each site was >50 mm, at around 4 weeks after tumor cell injection. After 4 weeks treatment, tumor progression in the CycD1b siRNA-treated group was significantly suppressed (Fig. 4b), which clearly demonstrated the inhibitory effect of CycD1b siRNA on established tumors. In addition, western blotting and CycD1b immunohistochemistry showed that CycD1b protein levels were decreased in most of the CycD1b siRNA-treated tumors (Fig. 4c,d). Indeed, considerably fewer cells overexpressing CycD1b were seen in tumors treated with CycD1b siRNA compared with those treated with PBS or control siRNA (Luc siRNA; Fig. 4d). Furthermore, the TUNEL assay revealed that there were more CycD1b apoptotic cells in the CycD1b siRNA-treated tumors (Fig. 4e), indicating that CycD1b depletion was able to induce apoptosis in vivo, leading to tumor suppression. We also tested the inhibitory efficacy of CycD1b siRNA using the tumorigenicity assay on another CycD1b-overexpressing cell line, namely SK-BR3 cells, as well as on T47D cells, a cell line in which CycD1b expression is low. Treatment with CycD1b siRNA markedly inhibited the tumor growth of SK-BR3 cells, but only exerted a mild suppressive effect on the growth of T47D cells in mice (Fig. 4f). Together, these results show that treatment with CycD1b siRNA can inhibit the growth of CycD1b-overexpressing breast tumor cells in vivo, indicating that CycD1b siRNA may serve as a novel therapeutic agent for the treatment of breast cancer in which CycD1b is overexpressed.

Figure 4.

 Tumor growth in vivo is inhibited by cyclin D1b (CycD1b) siRNA. (a) Growth of MDA-MB436 tumor xenografts is suppressed by CycD1b siRNA treatment. The red arrows indicated the days when mice were administered with breast cancer cells. Mice were killed on Day 40 after tumor cell injection and tumor samples were collected. Representative tumor samples from each group are shown on the right panel. Tumors from CycD1b siRNA-treated groups were much than those treated with mock or luciferase (Luc) siRNA. Data are the mean ± SD (= 3). *= 0.05 compared with mock transfection. (inline image), mock; (inline image), Luc siRNA; (inline image), CycD1b siRNA. (b) Growth inhibition of established tumor xenografts by CycD1b siRNA. MDA-MB436 cells were injected into two mammary glands of each nude mouse. When the tumors reached approximately 50 mm3 (around 4 weeks after tumor cell injection), mice were randomly treated with weekly intratumoral injections of mock or with Luc or CycD1b siRNA (10 mg). Mice were killed on Day 56 after cell injection and tumor samples were collected. Representative tumor samples from each group are shown on the right panel. Data are the mean ± SD (= 3). *= 0.05 compared with mock transfection. (inline image), mock; (inline image), Luc siRNA; (inline image), CycD1b siRNA. (c) Assessment of reduced CycD1b protein levels in CycD1b siRNA-treated tumors. MDA-MB436 tumors were collected 2 days after the last treatment with mock (M), Luc, or CycD1b siRNA and analyzed for CycD1b protein levels by western blotting. (d) Immunostaining confirmation of reduced CycD1b expression following CycD1b siRNA treatment. MDA-MB436 tumors were sectioned and analyzed for CycD1b expression using anti-CycD1b immunohistochemical staining (left panels). Data are summarized in the right-hand graph. Data are the mean ± SD (= 3). *= 0.01 compared with mock transfection. (e) Induction of apoptosis in CycD1b siRNA-treated tumors, as determined by TUNEL assay, with data summarized in the graph. Apoptotic cells stained brown. Data are the mean ± SD (= 3). *= 0.01 compared with mock transfection. (f) Tumor growth of SK-BR3, but not T47D, cells was significantly inhibited by CycD1b siRNA treatment in mice. From Day 14 after implantation of SK-BR3 or T47D cells, mice were treated weekly with indicated siRNA (10 μg) or liposome alone (mock). (inline image), mock; (inline image), Luc siRNA; (inline image), CycD1b siRNA. Data are the mean ± SD (= 3). *= 0.05 compared with mock transfection.

Synergistic inhibitory effects are achieved in vivo with a combination of CycD1b siRNA and Dox.  To broaden any potential clinical applications, we evaluated any synergistic effects on breast cancer cell growth of a combination of CycD1b siRNA and Dox (50 ng/mL), a chemotherapeutic drug commonly used to treat breast cancer patients.(36,37) In this series of experiments, SK-BR3 and MDA-MB436 cells were treated with CycD1b siRNA or Dox alone or in combination. The combination of CycD1b siRNA and Dox exhibited a much stronger cell-killing effect than CycD1b siRNA or Dox alone in both cell lines (Fig. 5a). Alone, 50 ng/mL Dox had no apparent effect on the expression of CycD1b in SK-BR3 and MDA-MB436 cells (Fig. 5b). To determine whether this inhibitory effect was synergistic, we calculated the CI using the CI-isobologram method developed by Chou.(34,35) As shown in Fig. 5(c), CI values at IC90 and IC95 were <1 for MDA-MB436 and SK-BR3 cells, indicating that there was a synergistic inhibitory effect on these cancer cells of the combination of CycD1b siRNA and Dox.

Figure 5.

 Enhanced inhibitory effects following combination treatment with cyclin D1b (cycD1b) siRNA plus doxorubicin (Dox). (a) Sensitization of breast cancer cells to Dox by cycD1b siRNA treatment. Cell viability was determined by the MTT assay after 8 days of the indicated treatment. Data are the mean ± SD (= 3). *= 0.05 compared with mock transfection. (b) Western blot analysis of the effects of Dox (50 ng/mL) on CycD1b levels in SK-BR3 and MDA-MB436 cells. (c) Synergistic cytotoxicity following combination treatment with cycD1b siRNA and Dox. SK-BR3 and MDA-MB436 cells were treated with different concentration of siRNA (0–0.1 mg/well in six-well plates) and split into 96-well plates. After 8 days incubation with different concentrations of Dox (0–100 ng/mL), cytotoxicity was determined using the MTT assay. The combination index (CI) was calculated as described in the Materials and Methods. Data show the CI value at 80%, 90%, and 95% cells killed. Data are the mean ± SD (= 3). (d) Effects of CycD1b siRNA in combination with Dox on breast tumor growth in mice. As described in the Materials and Methods, mice were treated for 4 weeks from Day 14 after MDA-MB436 cell injection with weekly injections of CycD1b siRNA (10 mg/injection; inline image), luciferase siRNA (inline image), liposome alone (mock; inline image), Dox alone (2 mg/kg, i.p.; inline image), or CycD1b siRNA + Dox (inline image). The effects of treatment on tumor growth were determined by measuring tumor volume. Data are the mean ± SD. *= 0.05, **= 0.01 compared with mock after four administrations.

We also investigated the therapeutic effects of this combined treatment in mice. Approximately 14 days after inoculation of MDA-MB436 cells, mice were divided into five groups (= 14 in each group). The different groups of mice were treated with liposome alone (mock), Luc siRNA, CycD1b siRNA, Dox alone, or a combination of Dox+CycD1b siRNA, as described in the Materials and Methods. Treatment with a combination of CycD1b siRNA and Dox was able to suppress the tumor growth of MDA-MB436 cells to a much greater degree than CycD1b siRNA oligonucleotides or Dox alone (Fig. 5d). These findings indicate that combined treatment with CycD1b siRNA plus Dox may provide a novel therapeutic option for breast cancer in which there is CycD1b overexpression.

Discussion

In many types of cancer, CycD1 is aberrantly expressed.(15,38–40) Indeed, high levels of the two isoforms of CycD1 are associated with a poor prognosis for breast cancer patients.(16,41,42) By using CycD1b siRNA to target CycD1b overexpression and assessing breast cancer growth in nude mice, we have shown that CycD1b siRNA effectively inhibits CycD1b overexpression. Tumorigenesis results from a disturbances of cell cycle progression and/or programmed apoptosis, which are perceived as reciprocal processes. Depletion of CycD1b promoted apoptosis of CycD1b-overexpressing cells and blocked their proliferation and transformation phenotype. We also demonstrated that CycD1b siRNA inhibited breast tumor growth in nude mice. Furthermore, we found that CycD1b siRNA synergistically enhanced the cell-killing effects of Dox in cell culture and this combination greatly suppressed tumor growth in mice. Thus, the present study clearly demonstrates the therapeutic potential of CycD1b siRNA for the treatment of CycD1b-overexpressing breast cancer, including triple-negative breast cancer. Our results also indicate that CycD1b, which is overexpressed in breast cancer, may serve as a novel and effective therapeutic target.

High levels of CycD1 are correlated with triple-negative breast cancers.(43,44) Indeed, we found that the overexpression of one isoform of CycD1, namely CycD1b, is correlated with triple-negative (ER, PR, and HER2 negative) basal-like breast cancers, which lack specific therapeutic targets. In the present study, two of three CycD1b-overexpressing cell lines used are triple-negative cancer cells (MDA-MB436 and MDA-MB157).(45,46) Of note, CycD1b siRNA alone or in combination with Dox effectively inhibited the growth of these cancer cells both in vitro and in vivo in mice. Thus, this combination therapy may be an effective treatment option for triple-negative breast cancers because, as yet, there are no specific treatment guidelines for these types of cancers, which appear to be very metastatic and have a poor prognosis.

Our results demonstrate that CycD1b siRNA exerts robust antitumor activity by promoting apoptosis and inhibiting DNA replication. The induced apoptosis is only observed in CycD1b-overexpressing cells (SK-BR3, MDA-MB157, and MDA-MB436) and not in cells in which CycD1b expression is low (MDA-MB453 and T47D). This specificity of action should increase the therapeutic index of siRNA-based therapies for CycD1b-overexpressing cancers. It is unclear how CycD1b depletion triggers apoptosis in CycD1b-overexpressing cells. However, it is well known that tumor cells are highly dependent on activated oncogenes for their survival and/or proliferation, a phenomenon called “oncogene addiction”. Inactivation of these activated oncogenes results in apoptosis and an antitumor action. The most convincing evidence for “oncogene addiction” comes from the examples of therapeutic efficacy of antibodies or drugs that target biomarkers in human cancers, such as the antibody trastuzumab (Herceptin, Genentech Inc., South San Francisco, CA, USA), targeting the receptor tyrosine kinase HER2/NEU in human breast cancer. The results of the experiments in the present study support the concept of “CycD1b addiction” in CycD1b-overexpressing cancers; that is, these cancers rely heavily on the activity of CycD1b for continued cell proliferation and survival. Importantly, in addition to promoting apoptosis, depletion of CycD1b by its siRNA sensitizes CycD1b-overexpressing breast cancer cells to the cell-killing effect of Dox. Doxorubicin is a chemotherapeutic drug that targets S-phase cells by inhibiting topoisomerase II.(47) Flow cytometry data from the present study demonstrate that depletion of CycD1b does not lead to accumulation of cells in the S phase, suggesting that sensitization to Dox is effected via mechanisms other than S phase accumulation. Interestingly, it has been shown that cell survival pathways, such as NF-κB, Akt, and the Bcl-2 family, can be activated to antagonize the cytotoxic effects of Dox.(48–51) Blockade of these pathways sensitizes breast cancer cells to Dox.(52) Therefore, it will be interesting to investigate in future studies whether CycD1b overexpression may enhance survival pathways and therefore increase resistance to Dox.

An effective delivery system is required for the practical application of siRNA in vivo therapy. In the present study, CycD1b siRNA was delivered by intratumoral injections of liposomes and this route of administration was found to be effective for the inhibition of CycD1b expression in vivo, resulting in tumor suppression. However, the CycD1b siRNA used in the present study was not chemically modified and so has relatively limited in vivo stability and membrane permeability. To apply CycD1b siRNA therapy in the clinical setting, it would be of considerable benefit if the siRNA could be modified to reduce its sensitivity to nucleases and to increase its cellular uptake. Various approaches have recently been used to chemically modify siRNA to increase its nuclease resistance and intracellular uptake.(53–56) It will be interesting to test the effects of these chemical modifications on the CycD1b siRNA, particularly with respect to improvements in therapeutic efficacy, especially for systemic treatment in mice. In addition to chemical modifications of siRNA, novel delivery systems can be used to improve the stability and efficacy of CycD1b siRNA,(57–59) such as carrying siRNA by targeted nanoparticles, and to thus further improve the therapeutic efficacy of CycD1b siRNA.

Acknowledgments

This work was supported by grants from the Science and Technology Commission of Shanghai Municipality (J-50208) and Shanghai Natural Science Funding (11ZR1422900).

Disclosure Statement

The authors have no conflict of interest.

Ancillary