Indolo-pyrido-isoquinolin based alkaloid inhibits growth, invasion and migration of breast cancer cells via activation of p53-miR34a axis

Abstract

The tumor suppressor p53 plays a critical role in suppressing cancer growth and progression and is an attractive target for the development of new targeted therapies. We synthesized several indolo-pyrido-isoquinolin based alkaloids to activate p53 function and examined their therapeutic efficacy using NCI-60 screening. Here, we provide molecular evidence that one of these compounds, 11-methoxy-2,3,4,13-tetrahydro-1H-indolo[2′,3′:3,4]pyrido[1,2-b]isoquinolin-6-ylium-bromide (termed P18 or NSC-768219) inhibits growth and clonogenic potential of cancer cells. P18 treatment results in downregulation of mesenchymal markers and concurrent upregulation of epithelial markers as well as inhibition of migration and invasion. Experimental epithelial–mesenchymal-transition (EMT) induced by exposure to TGFβ/TNFα is also completely reversed by P18. Importantly, P18 also inhibits mammosphere-formation along with a reduction in the expression of stemness factors, Oct4, Nanog and Sox2. We show that P18 induces expression, phosphorylation and accumulation of p53 in cancer cells. P18-mediated induction of p53 leads to increased nuclear localization and elevated expression of p53 target genes. Using isogenic cancer cells differing only in p53 status, we show that p53 plays an important role in P18-mediated alteration of mesenchymal and epithelial genes, inhibition of migration and invasion of cancer cells. Furthermore, P18 increases miR-34a expression in p53-dependent manner and miR-34a is integral for P18-mediated inhibition of growth, invasion and mammosphere-formation. miR-34a mimics potentiate P18 efficacy while miR-34a antagomirs antagonize P18. Collectively, these data provide evidence that P18 may represent a promising therapeutic strategy for the inhibition of growth and progression of breast cancer and p53-miR-34a axis is important for P18 function.

1 Introduction

The tumor suppressor p53 is a powerful transcription factor known to regulate various biological processes such as cell growth, apoptosis, angiogenesis, invasion and migration via transcriptional modulation of multiple genes (Duffy et al., 2014; Vogelstein et al., 2000). Owing to its critical role in tumor suppression, it is not surprising that p53 gene is mutated and functionally inactivated at an extremely high frequency in all human tumors (Feki and Irminger-Finger, 2004; Hainaut and Hollstein, 2000). Also, in tumors carrying wild-type p53, the function of wild-type p53 is inhibited by its primary endogenous cellular inhibitor, the murine double minute 2 (MDM2) oncoprotein (called HDM2 in humans) (Wu et al., 1993). MDM2 and p53 form an autoregulatory feedback loop where MDM2 binds to the amino-terminal transactivation domain of p53 leading to inhibition of its transcriptional activity and nuclear export. p53, on the other hand, transcriptionally up-regulates MDM2 expression resulting in an inhibition of p53 activity (Chen et al., 1993; Wu et al., 1993). In addition, MDM2 being an E3 ubiquitin ligase promotes proteasome-mediated p53 degradation (Freedman et al., 1999). Hence, restoration of p53 function in tumors by blocking MDM2–p53 interaction or by inhibiting the ubiquitin-ligase (E3) activity of MDM2 is considered an effective strategy for treating cancer. In fact, many small molecules capable of reactivating p53 are currently undergoing preclinical testing and several compounds are already in clinical trials (Duffy et al., 2014; Selivanova, 2014). Some of these compounds exclusively inhibit MDM2 mediated p53 ubiquitylation while others inhibit MDM2–p53 interaction as well as MDM2 auto-ubiquitylation.

In recent years, active constitutive agents in natural products have shown efficacy as potential cancer preventive as well as therapeutic agents. In fact, more than half of the chemotherapeutic agents in clinic are derived from natural products (Newman et al., 2003). Screening of >140,000 natural products and natural products extracts in an electrochemiluminescent assay system discovered that sempervirine inhibits MDM2-mediated p53 ubiquitylation as well as MDM2 auto-ubiquitylation. Importantly, sempervirine preferentially induced p53-mediated cell death in transformed cells while exhibiting little toxicity in normal cells (Sasiela et al., 2008). Sempervirine is a natural indolo[2,3-a]quinolizine based alkaloid belonging to the group of sempervirines from the beta-carboline class (Beljanski and Beljanski, 1982). Sempervirine treatment has suppressive effects on tumor growth in BALB/C mice inoculated with transplantable YC8 lymphoma ascites cells and Swiss mice bearing Ehrlich ascites carcinoma cells (Beljanski and Beljanski, 1986). The discovery that sempervirine modulates p53–MDM2 interaction provided the underlying mechanism of action for this long known chemoprotective agent.

While many research groups have pursued the design of peptides and peptidomimetics to inhibit p53 and MDM2 interactions, the biological significance of sempervirine has not been studied. To better understand the importance of indole and quinolizine moieties in the overall biological activity of sempervirine, we modified the original structure and synthesized twenty analogs which were further evaluated for their biological efficacy. In the present study, we specifically investigated the potential of these analogs to inhibit growth and clonogenic potential of breast cancer cells. Our study identified that the analog containing both indole and quinolizine moieties and a methoxy substituent (termed P18) was most effective. Not only does P18 inhibit growth and clonogenic potential of cancer cells, it also effectively inhibits migration, and invasion potential of breast cancer cells. We provide strong evidence that P18 inhibits experimental epithelial–mesenchymal-transition (EMT) as well as mammosphere formation along with alterations of epithelial-mesenchymal genes and inhibition of stemness factors. We show that P18 induces expression, phosphorylation and accumulation of p53 in cancer cells. A key role of p53 is proposed in P18 function as we show that p53 is important for P18-mediated alteration of mesenchymal and epithelial genes, inhibition of migration and invasion of cancer cells. We found that P18 also increases miR-34a expression in p53-dependent manner. Inhibition of growth, invasion and mammosphere-formation by P18 is further enhanced by miR-34a mimic while miR-34a antagomir antagonizes P18 showing integral role of miR-34a in P18 function. Our studies demonstrate P18 may represent a promising therapeutic strategy for inhibiting growth, EMT, invasion and migration of breast cancer.

2 Materials and methods 2.1 Synthesis and characterization of the sempervirine analogs

General procedure for quaternization: To a solution of the heterocycle (10.0 mmol) in ethanol (20 mL) or in acetone (20 mL), ethyl bromoacetate (15.0 mmol, 1.5 eq) was added at room temperature. The resulting reaction mixture was refluxed for 16 h. The progress of the reaction was monitored by TLC. The reaction mixture was cooled and concentrated under reduced pressure. The residue was crystallized in acetone (∼100 mL) to give the desired quaternary ammonium salt. General procedure for Westphal condensation reaction: To a solution of quaternary ammonium bromide derivative (1.0 mmol) in methanol (4 mL), diketo compound (1.2 mmol, 1.2 eq) and sodium methoxide (4.0 mmol, 4.0 eq) were added at room temperature. The resulting reaction mixture was heated in CEM-Microwave (Explorer) between 95 °C and 105 °C (Power 150 W and Pressure 250 psi) temperature for 60 min. The reaction mixture was acidified with acetic acid and concentrated under reduced pressure. The crude product was purified by column chromatography over Silica-gel, eluting with a mixture of methanol and dichloromethane. ¹H NMR (400 MHz, DMSO-d₆) δ: 1.90 (m, 4H), 2.99 (m, 2H), 3.15 (m, 2H), 3.95 (s, 3H), 7.03 (m, 1H), 7.21 (s, 1H), 8.26 (d, J = 8.8 Hz, 1H), 8.58 (d, J = 7.2 Hz, 1H), 8.66 (s, 1H), 8.82 (d, J = 7.2 Hz, 1H), 9.19 (s, 1H), 13.21 (s, 1H); MS (m/z): 303.25; m.p.: 207–212 °C.

2.2 NCI-60 screening

Analogs were submitted to the National Cancer Institute for screening. Details of the methodology for NCI 60 cell line screening are described at http://dtp.nci.nih.gov/branches/btb/ivclsp.html.9. Briefly, the panel is organized into nine subpanels representing diverse histologies: leukemia, melanoma, and cancers of lung, colon, kidney, ovary, breast, prostate, and central nervous system. The cells are grown in supplemented RPM1 1640 medium for 24 h. The test compounds were dissolved in DMSO and incubated with cells at five concentrations with 10-fold dilutions, the highest being 10⁻⁴ M and the others being 10⁻⁵, 10⁻⁶, 10⁻⁷, and 10⁻⁸ M. The assay is terminated by addition of cold trichloroacetic acid, and the cells are fixed and stained with sulforhodamine B. Bound stain is solubilized, and the absorbance is read on an automated plate reader. The cytostatic parameter 50% growth inhibition (GI₅₀) was calculated from time zero, control growth and the five concentration level absorbance. The cytotoxic parameter that is, inhibitory concentrations (LC₅₀) represent the average of two independent experiments. The in vitro screening is a two-stage process started with the evaluation of compound against the 60 human tumor cell lines with a single dose of 10.0 μM, which is done by following same protocol as for five dose screening. Only the compounds that show more than 60% of growth inhibition in at least 8 tumor cell lines are selected for further testing and the others were assumed inactive.

2.3 Cell culture and reagents

The human breast cancer cell lines, MCF7, HBL100, HMEC and MCF10A were obtained from the American Type Culture Collection (ATCC, Manassas, VA), resuscitated from early passage liquid nitrogen vapor stocks as needed and cultured according to supplier's instructions. Cell line authentication was done by analysis of known genetic markers or response (e.g. expression of estrogen receptor and p53 and estrogen responsiveness). Cells were cultured for less than 3 months before reinitiating cultures and were routinely inspected microscopically for stable phenotype. MCF10A is nontumorigenic and widely used as a representative normal mammary epithelial cell line. MCF10A was isolated from fibrocystic breast disease and spontaneously immortalized. HCT116 p53^−/− and HCT116 p53^+/+ cells were kindly provided by Dr. Bert Vogelstein (Johns Hopkins University, Baltimore, MD, USA). HCT116 p53^−/− and HCT116 p53^+/+ cell lines were cultured in McCoy's 5A medium (Gibco-BRL) containing 10% fetal bovine serum and antibiotics. For treatment, cells were seeded at a density of 1 × 10⁶/100-mm tissue culture dish and treated with p18 as indicated. We synthesized P18 following our previously published protocols (Rao et al., 2013). TGFβ was purchased from Calbiochem (Billerica, MA) and TNFα was obtained from Sigma–Aldrich (St. Louis, MO). Antibodies for Nanog (D73G4), Oct4 (2750), Sox2 (D6D9), phospho-p53-S15, phospho-p53-Thr18 and MDM2 were purchased from Cell Signaling. Antibodies for p53 and p21 were purchased from Santa Cruz biotechnology. Antibodies for p27 were procured from Invitrogen. Antibodies for β-Actin were purchased from Sigma. Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) staining was performed using TUNEL apoptosis detection kit (EMD Millipore).

2.4 Cell viability assay

Cell viability assay was performed by estimating reduction of XTT (2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxyanilide), using a commercially available kit (Roche Applied Science, Indianapolis, IN) following manufacturer's instructions. MCF7, HBL-100, MCF10A and HMEC cells were plated in 96 well plates at an initial density of 4 × 10³ cells/well for 24 h followed by treatment with P18 as indicated and the medium was replaced with fresh medium containing treatments every 3 days. XTT labeling reagent was added to each culture well to attain a final concentration of 0.3 mg/mL. After 4 h exposure at 37 °C, absorbance was measured at 450 and 690 nm using a 96 well plate reader (SPECTRAmax PLUS, Molecular Devices, CA).

2.5 Clonogenicity assay

Colony formation assay was performed following our published protocol. Breast cancer cells (single-cell suspension) were plated in 12-well plates at a density of 250 cells per well. Cells were allowed to adhere for 24 h followed by treatment with P18 and the medium was replaced with fresh medium containing treatments every 3 days. After a 10-day treatment period, the medium was removed and colonies were stained with crystal violet (0.1% in 20% methanol). Colony numbers were assessed visually and colonies containing >50 normal-appearing cells were counted. Pictures were taken using a digital camera.

2.6 RNA isolation, miR, transfection and RT-PCR

For RNA isolation and RT-PCR, total cellular RNA was extracted using the TRIzol Reagent (Life Technologies, Inc., Rockville, MD). RT-PCR was performed using specific sense and antisense PCR primers. Cells were transfected with miR-34a mimic, antagomir or control-miR (Applied Biosystems, Ambion, Austin, TX) using Fugene transfection reagent (Promega Corporation, Madison, WI). For qRT-PCR detection of miR-34a, miRNA-specific RT-primers (assay IDs: hsa-miR-34a, 000426), TaqMan miRNA Assay (Applied Biosystems, Ambion, Austin, TX) and Platinum Taq Polymerase Reagents (Invitrogen, Grand Island, NY) were used. Data were calculated by using the standard ΔΔCt method and microRNA expression was represented as fold-difference of each treatment vs. vehicle-treated control. Statistics was performed by using one-way ANOVA and Student's t-test post-hoc analysis. Statistical significance was accepted when p was <0.05. For RNA interference, cells were transfected at 50% confluency with 100 nM of control siRNA or p53 siRNA (SignalSilence, Cell Signaling Technology) using Oligofectamine (Thermo Fisher Scientific).

2.7 Scratch-migration assay

To perform migration assays; cells were plated into the 6-well cell culture plate. Cells were allowed to grow in DMEM containing 10% FBS to confluence, and then were washed with serum-free medium and serum starved for 16 h. A 1-mm wide scratch was made across the cell layer using a sterile pipette tip. Plates were photographed immediately after scratching. Cells were treated with P18 and plates were photographed after 8 h, 24 h, 48 h and 72 h at the identical location of the initial image.

2.8 Spheroid migration assay

MCF7 and HBL100 cells (1.5 × 10⁴) were seeded in 0.5% agar-coated plates and cultured on an orbital shaker (100 rpm) for 48 h in a humidified atmosphere containing 5% CO₂ at 37 °C. Intact tumor spheroids were selected and transferred to six-well plates. The spheroids were treated with P18 as indicated. After 48 h of incubation, spheroids were fixed with 10% buffered formalin in PBS and stained with crystal violet. The migration of cells from spheroids was observed under light microscope.

2.9 Invasion assay

For an in vitro model system for metastasis, a Matrigel-invasion assay was performed using a Matrigel invasion chamber from BD Biocoat Cellware (San Jose, CA). The slides were coded to prevent counting bias, and the number of invaded cells on representative sections of each membrane were counted under light microscope. The number of invaded cells for each experimental sample represents the average of triplicate wells.

2.10 Immunofluorescence and confocal imaging

Breast cancer cells (5 × 10⁵ cells/well) were plated in 4-well chamber slides (Nunc, Rochester, NY) followed by treatment with P18 as indicated and subjected to immunofluorescence analysis. Fixed and immunofluorescently stained cells were imaged using a Zeiss LSM510 Meta (Zeiss) laser scanning confocal system configured to a Zeiss Axioplan 2 upright microscope. All experiments were performed multiple times using independent biological replicates.

2.11 Mammosphere assays

Mammosphere assays were performed as previously described and spheres (>50 μm) were counted (Shaw et al., 2012). Single cells were plated in ultra-low attachment plates (Corning) at a density 500 to 5000 cells/well in serum-free mammary epithelium basal medium (Lonza) supplemented with 1% penicillin/streptomycin, B27 (1:50, Invitrogen-Life Technologies), 5 μg/mL insulin, 1 μg/mL hydrocortisone (Sigma), 20 ng/mL EGF (R&D Systems), 20 ng/mL basic fibroblast growth factor (Stem Cell), and 2-mercaptoethanol. BITC was added as indicated to the media. The mammospheres were counted under an inverted microscope.

2.12 Western blotting

Whole cell lysate was prepared by scraping breast cancer cells in 250 μL of ice cold modified RIPA buffer. Equal amount of lysate protein was resolved on sodium-dodecyl sulfate polyacrylamide gel, transferred to nitrocellulose membrane, and western blot analysis was performed. Immunodetection was performed using enhanced chemiluminescence (ECL system, Amersham Pharmacia Biotech Inc., Arlington Heights, IL) according to manufacturer's instructions.

2.13 Statistical analysis

All experiments were performed multiple times. Statistical analysis was performed using Microsoft Excel software. Significant differences were analyzed using Student's t test and two-tailed distribution. Results were considered to be statistically significant if p < 0.05. Results were expressed as mean ± SE between triplicate experiments performed thrice.

3 Results 3.1 P18 (NSC-768219 or 11-methoxy-2,3,4,13-tetrahydro-1H-indolo [2′,3′:3,4] pyrido [1,2-b] isoquinolin-6-ylium bromide) inhibits growth and clonogenic potential of cancer cells

Sempervirine is an alkaloid bearing two basic moieties, i.e. indole and quinolizine. To better understand the importance of each moiety in the overall biological activity of sempervirine, we modified the original structure and synthesized a host of analogs listed in Figure 1A. We first examined the potential of sempervirine analogs as multitargeting agents by screening in the wide range of cancer cell lines in the NCI-60 panel. The structural changes included removal of indole subunit (P3–5, P10–17, and P19), removal of quinolizine subunit (P6–9), and introducing new substrate on the indole ring of the parent compound. The anticancer screening of these analogs revealed few important features, i.e. (i) the quinolizine unit is important, as its removal resulted in nearly inactive compounds (P3,5,10,16,17) and showed some activity when phenyl rings were introduced as additional substitution on the ring. (ii) Similarly, indole plays an important role and contributes to the biological activity of sempervirine. Complete removal of indole with no changes on the quinolizine unit makes the compounds very inactive (P6). Addition of phenyl substituents on the pyridine ring of indole improves activity for some of the cancer cell lines. (iii) However, the most effective modification was obtained when both indole and quinolizine were intact and a methoxy substituent was introduced (P18). This compound was very active against a host of cancer panels (Figure 1B). The heat map in Figure 1B summarizes the patterns of cytostatic behavior with high intensity values (red) indicate higher activity (lower GI₅₀), and lower intensity values (blue) indicate lower activity (higher GI₅₀) for all analogs. P18 (at single dose of 10 μM) shows effective inhibition of cell growth in majority of cell lines. A high LC₅₀ value compare to GI₅₀ suggest wider therapeutic window for further development. Higher efficacy of P18 against breast cancer cell line panel motivated us to further investigate the biological functions and mechanism of action of P18 in breast cancer cells. Breast cancer cell lines, MCF7 and HBL-100 were treated with various concentrations ranging from 1 μM to 10 μM or 50 μM P18 and subjected to XTT cell proliferation assay (Figure 2A) and clonogenicity assay (Figure 2B). Dose-dependent and statistically significant inhibition of cell growth was observed in the presence of P18. Treatment with 2.5 μM and 5 μM P18 resulted in ∼50% inhibition in growth and clonogenicity, whereas higher concentrations were more inhibitory (Figure 2A,B). P18 did not affect growth of MCF10A and HMEC (Human Mammary Epithelial Cells) cells (Figure 2A). P18 treatment induces apoptotic cell death in MCF7 and HBL100 cells (Figure 2C). Next, breast cancer cells were treated with equal concentrations of sempervirine and P18 followed by clonogenicity assay. P18 is significantly more effective than sempervirine in inhibiting breast cancer growth (Supplementary Figure 1).

Evaluation of sempervirine analogs including P18 in NCI-60 screening. (A) Sempervirine analogs were synthesized. IUPAC nomenclature of analogs studied. (B) Analogs were evaluated in the NCI-60 screen. Heat map shows the log10 GI50 values for each sempervirine analog in the NCI-60 screen, where high intensity (red) cells indicate high activity and low intensity (blue) cells indicate low activity.

P18 inhibits breast cancer growth, modulates epithelial and mesenchymal markers and inhibits migration and invasion of breast cancer cells. (A) MCF7, HBL100, MCF10A and HMEC cells were treated with various concentration of P18 for 24, 48 and 72 h as indicated and subjected to XTT cell proliferation assay. *p [less than] 0.05 compared with untreated controls. Vehicle-treated cells are denoted with C. (B) Clonogenicity of breast cancer cells treated with various concentrations of P18 (as indicated). Representative images are shown. (C) TUNEL staining of breast cancer cells treated with P18. Bar graph shows number of TUNEL positive cells. *p [less than] 0.001 compared with untreated controls (C). (D) Invasion potential of MCF7 cells treated with 5 μM P18 was examined using Matrigel-invasion assay. (E) Breast cancer cells were treated with 5 and 10 μM P18 and subjected to spheroid migration assay. (F) Breast cancer cells were treated with 5 μM P18 followed by scratch-migration assay. (G) MCF-7 cells were treated with 5 μM P18 for indicated time intervals. Total RNA was isolated and expression level of epithelial (E-cadherin and CK-18) marker genes, mesenchymal (Fibronectin and Vimentin) marker genes and EMT-related transcription factors (Snail and Zeb1) was analyzed. Actin was used as control.

3.2 P18 inhibits invasion, migration potential and modulates the expression of mesenchymal and epithelial markers of EMT in breast cancer cells

Cancer cells invade through basement membrane and migrate to distant sites to form metastatic lesions during metastatic progression. Upon examining the effect of P18 on invasion potential of breast cancer cells using Matrigel-invasion assay, we found that treatment with 5 μM P18 inhibited invasion of breast cancer cells through Matrigel in comparison to control cells (Figure 2D). Scratch-migration and spheroid-migration assays were utilized to investigate if P18 impacted migration capacity of breast cancer cells. Significant migration of MCF7 and HBL-100 cells from the spheroids was observed in vehicle-treated conditions while P18 treatment resulted in inhibition of migration (Figure 2E). P18 treatment reduced migration of breast cancer cells in a scratch-migration or wound-healing assay (Figure 2F). Epithelial to mesenchymal transition (EMT) of cancer cells of epithelial origin is an essential step that precedes the induction of motility and invasive potential during metastatic progression of malignant cells (Dave et al., 2012). Since P18 inhibited invasion and migration of MCF7 and HBL-100 cells, we aimed to examine whether P18 treatment influenced EMT in breast cancer cells. It is known that biochemical hallmarks of EMT-reversal include gain of expression of epithelial marker proteins such as E-cadherin, occludin and cytokeratin-18 with a concomitant decrease in mesenchymal marker (e.g. vimentin, fibronectin, and N-cadherin) expression (Dave et al., 2012). It was interesting to note that treatment of breast cancer cells with P18 resulted in inhibition of fibronectin and vimentin expression and an accompanied increase in the expression of E-cadherin and CK-18 along with inhibition of EMT-related transcription factors, Snail and Zeb1 (Figure 2G).

3.3 Inhibition of epithelial-mesenchymal transition in TGFβ/TNFα-treated mammary epithelial cells by P18

EMT can be experimentally induced in some cell types by treating with a combination of TGFβ and TNFα (Bates and Mercurio, 2003; Sehrawat and Singh, 2011; Willis et al., 2005). Morphologic changes from an epithelial-like to mesenchymal-like appearance were induced when MCF10A normal mammary epithelial cells were treated with TGFβ/TNFα. TGFβ/TNFα-treated MCF10A cells exhibited mesenchymal phenotype acquiring spindle-like appearance and increased intracellular separation signifying loss of intercellular adhesion while vehicle treated MCF10A cells showed round, well-packed cobblestone appearance, a characteristic of epithelial cells (Figure 3A). We examined the possibility that P18 treatment is able to block TGFβ/TNFα-induced EMT in MCF10A cells. Interestingly, P18 treatment prevented the morphologic transition from an epithelial-like to mesenchymal-like appearance caused by TGFβ/TNFα treatment. P18 alone did not affect the morphology of MCF10A (Figure 3A). Our results showed that TGFβ/TNFα-treated MCF10A cells exhibited elevated levels of mesenchymal markers (fibronectin and vimentin) and EMT-related transcription factors (Snail and Zeb1). P18 treatment blocked TGFβ/TNFα-induced modulation of mesenchymal markers and EMT-related transcription factors (Figure 3B). Immunofluorescence analyses showed that TGFβ/TNFα treatment induced nuclear localization of Snail, Slug and Zeb1 which can be inhibited with P18 treatment (Figure 3C). All these findings confirm the potential of P18 as a novel EMT-inhibitor using an experimental system involving TGFβ/TNFα and MCF10A cells.

P18 inhibits TGFβ/TNFα-induced epithelial-mesenchymal transition in mammary epithelial cells. (A) MCF10A cells were treated with vehicle control (control), TGFβ + TNFα (10 ng/mL of each, TT), 5 μM P18 or TGFβ + TNFα+P18 for 72 h. Morphological changes associated with EMT are shown in phase-contrast images. The presence of spindle-shaped cells, increased intercellular separation and pseudopodia were noted in TGFβ + TNFα-treated cells but not in P18 treated or TGFβ + TNFα + P18-treated cells. (B) MCF10A cells were treated as in A, total RNA was isolated and expression of mesenchymal markers (fibronectin, vimentin) and EMT-related transcription factors (SNAIL, ZEB2) was analyzed. Actin was included as control. (C) MCF10A cells were treated as in A, and subjected to immunofluorescence analysis of Snail, Slug and Zeb1 (Magnification 1000×).

3.4 P18 reduces mammosphere formation and inhibits stemness markers in breast cancer cells

Previous studies have shown that several EMT-inducing factors bestow cells with stem-like characteristics and facilitate an increase in the subpopulation of CSC (cancer stem cells) (Mani et al., 2008; Morel et al., 2008; Santisteban et al., 2009). Utilizing mammosphere assay that relies on the unique property of breast cancer cells with stem-like potential to form large, round, unattached floating spheroid colonies (termed mammosphere), we show that P18 treatment reduces mammosphere formation (Figure 4A). Owing to the association of induced pluripotent stem cell (iPSC) markers, Nanog, Oct4 and Sox2 with self-renewal and maintenance of stem cell fate, we examined whether iPSC markers are affected by P18 treatment. We investigated the expression of Oct4, Nanog and Sox2 in P18-treated breast cancer cells and found that P18 treatment inhibited the expression of pluripotency genes and proteins in comparison to control-treated cells (Figure 4B,C). P18 does not inhibit expression of other early response transcription factors (Supplementary Figure 2). Immunofluorescence analyses showed that TGFβ/TNFα treatment induced nuclear localization of Oct4 which can be inhibited with P18 treatment (Figure 4D). Together, these results show that P18 treatment results in effective inhibition of mammosphere formation and iPSC markers of breast cancer cells.

P18 suppresses mammosphere-formation potential and acquisition of stem-like properties in breast cancer cells. (A) MCF7 and HBL-100 cells were treated with 5 and 10 μM P18 and subjected to mammosphere formation. Vehicle treated cells are denoted as (C). The graph shows the number of mammospheres. (B) MCF7 cells were treated with 5 μM P18 for indicated time intervals, total RNA was isolated and expression level of stemness marker (Oct4, Nanog and Sox2) genes was analyzed. Actin was used as control. (C) MCF7 and HBL100 cells were treated with 5 μM P18 for indicated time intervals, total protein was isolated and expression level of stemness marker (Oct4, Nanog and Sox2) genes was analyzed using immunoblot analysis. Actin was used as control. (D) MCF7 cells were treated with vehicle control (control), TGFβ + TNFα (10 ng/mL of each, TT), 5 μM P18 or TGFβ + TNFα+P18 and subjected to immunofluorescence analysis of Oct4 (Magnification 1000×).

3.5 P18 induces expression and nuclear localization of p53 and inhibits EMT, migration and invasion in a p53-dependent manner

Tumor suppressor p53 is known to inhibit cancer progression via multiple biological processes including induction of growth arrest, senescence and apoptosis; modulation of tumor stroma and angiogenesis; blockade of invasion, migration and metastasis (Duffy et al., 2014; Hainaut and Hollstein, 2000). We show that P18 treatment increases p53 expression in breast cancer cells (Figure 5A,B). Posttranslational modifications, including phosphorylation, play an important role in stabilization and activation of p53 (Brooks and Gu, 2003; Woods and Vousden, 2001). We found that P18 treatment significantly stimulated phosphorylation of p53 at Ser15 in breast cancer cells (Figure 5B). Furthermore, P18 induced nuclear translocation of p53 (Figure 5C) and increased the expression of p53-target genes-p21 and p27 in breast cancer cells (Figure 5A,B). The tumor suppressor p53 and MDM2 are linked through a negative autoregulatory loop. MDM2 itself is the product of p53-inducible gene which maintains p53 at a low levels (Haupt et al., 1997). It is known that phosphorylation of p53 at Thr18 abrogates p53–MDM2 binding leading to an increase in p53 stability (Schon et al., 2002). We found that P18 treatment leads to an increase in MDM2 expression but importantly, it also increases p53-Thr18 phosphorylation (Figure 5D). These results indicate that P18 can inhibit p53–MDM2 interaction via inducing p53 phosphorylation at Thr18. Next we investigated the importance of p53 in P18-mediated modulation of EMT markers, migration and invasion of breast cancer cells. Towards this goal, HCT116 p53^−/− and HCT116 p53^+/+ cells (Figure 5E) were treated with various concentrations of P18 and analyzed for alteration of EMT markers. We found that P18 treatment decreased the expression of mesenchymal markers fibronectin and vimentin and increased epithelial marker E-cadherin expression in wild-type p53 HCT116 p53^+/+ cells. P18 did not inhibit the expression of mesenchymal markers in HCT116 p53^−/− cells. Also, P18 treatment significantly decreased the expression of EMT-related transcription factors Zeb1, Snail and Slug in HCT116 p53^+/+ cells whereas no reduction was observed in HCT116 p53^−/− cells (Figure 5F). Examining the functional importance of p53 in mediating inhibitory effects of P18 on migration and invasion of breast cancer cells, we found that P18 inhibited migration of HCT116 p53^+/+ cells in a scratch-migration assay while HCT116 p53^−/− cells exhibited extensive migration even in the presence of P18 (Figure 5G). Also, HCT116 p53^+/+ cells treated with P18 showed significant reduction in invasion through Matrigel in comparison to P18-treated HCT116 p53^−/− cells (Figure 5H). Next, the effect of P18 treatment on migration and invasion of MCF7 cells was examined in p53 silenced state. p53-silenced MCF7 cells exhibited no change in invasion as well as migration potential upon P18 treatment whereas control-si-MCF7 cells showed significant inhibition (Figure 5 I and J). These results show that anti-oncogenic effects of P18 are mediated via p53 activation.

P18 induces p53 expression, phosphorylation and increases nuclear localization of p53 in breast cancer cells. p53 plays an important role in mediating biological effects of P18. (A) MCF7 cells were treated with 5 μM P18 for indicated time-intervals, total RNA was isolated followed by RT-PCR analysis for p53, p27 and p21. Actin was used as control. (B) MCF7 cells were treated with 5 μM P18 for indicated time-intervals, total protein was isolated followed by immunoblot analysis for p53, p27 and p21. Actin was used as control. (C) MCF7 cells were treated with 5 μM of P18 followed by immunofluorescence microscopy against phosphorylated p53 at Serine 15 (p-p53-S15) (stained with Cy5, red). Nuclei were stained with DAPI (blue). (Magnification 1000×). (D) MCF7 cells were treated with 5 μM P18 for indicated time-intervals, total protein was isolated followed by immunoblot analysis for p53-Thr18 and MDM2. Actin was used as control. (E) Total protein was isolated from HCT116-p53+/+ and HCT116-p53−/− cells and immunoblotted for p53 expression. (F)Total RNA was isolated from HCT116-p53+/+ and HCT116-p53−/− cells treated with 5 and 10 μM of P18 as indicated and expression of mesenchymal markers (fibronectin, vimentin), epithelial marker (E-cadherin) and EMT-related transcription factors (Snail, Slug and Zeb1) was analyzed. Actin was included as control. (G) HCT116-p53+/+ and HCT116-p53−/− cells treated with 5 and 10 μM of P18 as indicated and subjected to scratch-migration assay. Bar graphs show fold change in migration. *p [less than] 0.01, compared with untreated controls. (H) HCT116-p53+/+ and HCT116-p53−/− cells treated with 5 μM of P18 as indicated and subjected to invasion assay. Bar-graph shows numbers of cells invaded through Matrigel. Representative pictures are shown. *p [less than] 0.05, compared with untreated controls. (I) MCF7 cells were transfected with control-si or p53-si as indicated, treated with 5 and 10 μM of P18 and subjected to invasion assay. *p [less than] 0.005, compared with untreated controls. (J) MCF7 cells were transfected with control-si or p53-si as indicated, treated with 5 and 10 μM of P18 and subjected to scratch-migration assay. *p [less than] 0.005, compared with untreated controls.

3.6 Evidence supporting a role of miR34a in mediating inhibition of growth, invasion and mammosphere formation by P18 in HCT116 p53^+/+ cells

It is well-recognized that microRNAs (miRNAs) play an important regulatory role in various biological and pathological processes including cancer growth, progression and metastasis. miRs are small (∼21-mer) regulatory RNA molecules known to mediate their regulatory effects by binding to the 3′untranslated regions (3′UTR) of specific mRNAs resulting in mRNA degradation or translational repression (Nicoloso et al., 2009). Functioning as oncogenes or tumor suppressors, miRNAs can either potentiate cancer progression or mediate growth inhibition (Tavazoie et al., 2008). One of the important tumor suppressor miRNA, known to be downregulated in aggressive breast cancer, is miR34a (Yang et al., 2013). MCF7 and HBL-100 cells treated with P18 exhibited a time-dependent increase in miR-34a expression (Figure 6A). We further investigated whether p53 plays any role in P18-mediated increase of miR-34a. MCF7-control-si (p53-positive) and MCF7-p53-si (p53-negative) cells were treated with P18 and expression of miR-34a was determined. Intriguingly, displaying a crucial role of p53, P18 treatment did not increase miR-34a expression in MCF7-p53-si cells while MCF7-control-si cells exhibited P18-induced miR-34a expression (Figure 6B). Also, HCT116 p53^+/+ cells showed increased levels of miR-34a upon P18 treatment whereas HCT116 p53^−/− cells exhibited no change (Supplementary Figure 3).

p53 plays an integral role in P18-mediated upregulation of miR-34a expression. miR-34a is integral for P18-mediated inhibition of growth, invasion and mammosphere formation. (A) TaqMan RT-PCR analysis of miR-34a expression in MCF-7 and HBL-100 cells treated with 5 and 10 μM P18 as indicated. C denotes vehicle-control. *p [less than] 0.05, compared with untreated controls. (B) MCF7 cells were transfected with control-si or p53-si as indicated, treated with 5 and 10 μM of P18 and total RNA was isolated followed by TaqMan RT-PCR analysis of miR-34a expression. C denotes vehicle-control. *p [less than] 0.05, compared with untreated controls. (C) MCF7 and HBL100 cells were transfected with miR-34a mimic or antagomir (inhibitor), treated with P18 as indicated followed by XTT cell proliferation assay. *p [less than] 0.001 compared with untreated controls. **p [less than] 0.005 compared with P18-treated cells. Vehicle-treated cells are denoted with C. (D) MCF7 and HBL100 cells were transfected with miR-34a mimic or antagomir (inhibitor), treated with P18 as indicated followed by Matrigel-invasion assay. *p [less than] 0.01 compared with untreated controls. **p [less than] 0.001 compared with P18-treated cells. Vehicle-treated cells are denoted with C. (E) MCF7 and HBL100 cells were transfected with miR-34a mimic or antagomir (inhibitor), treated with P18 as indicated followed by mammosphere assay. Data shows numbers of mammosphere. *p [less than] 0.05, compared with untreated controls. **p [less than] 0.01 compared with P18-treated cells.

To examine the functional relevance of miR-34a in P18-mediated inhibition of growth, invasion and mammosphere formation, we overexpressed miR-34a (in the form of a mimic molecule) or inhibited miR-34a (using antagomir) in MCF7 and HBL100 cells followed by treatment with vehicle or P18. As evident in Figure 6C–E, P18 significantly decreased growth, invasion and mammosphere formation in breast cancer cells which was further inhibited in the presence of miR-34a mimic. Inhibiting miR34a resulted in abrogation of the efficacy of P18 (Figure 6C–E). P18 treatment also inhibits expression of miR-34a target genes (Supplementary Figure 4). Together, these data provide evidence supporting an important role of miR-34 in P18-mediated inhibition of growth, invasion and mammosphere formation.

Collectively, the findings presented here suggest that P18 inhibits breast cancer growth, invasion, migration and mammosphere formation, and provide evidence for the involvement of p53 and miR-34a. We uncover a mechanism of P18 action through activation of miR-34a in a p53-dependent manner.

4 Discussion

A compelling body of evidence has put forth that p53 tumor suppressor is the most frequently inactivated gene in cancer. Studies utilizing various in vitro model systems and mouse models have established that reconstitution of p53 function inhibits growth and progression of established tumors (Martins et al., 2006; Ventura et al., 2007; Xue et al., 2007). Importantly, these studies show that established tumors are sensitive to p53 induction while normal cells are not significantly altered upon p53 re-establishment. Activation of p53 inhibits cancer growth and progression by affecting various biological processes such as, induction of growth arrest, apoptosis, inhibition of angiogenesis, blocking invasion and migration (Christophorou et al., 2005; Vogelstein et al., 2000; Vousden and Prives, 2009). Various strategies have been developed for the reconstitution of p53 function depending on the mechanisms leading to p53 inactivation. We present here the development of novel small-molecules based on the structure of sempervirine which is known to inhibit MDM2-mediated p53 ubiquitylation as well as MDM2 auto-ubiquitylation leading to p53 activation; and detailed studies of their anti-breast cancer activities and mechanism of action. All sempervirine analogs were screened in the wide range of cancer cell lines in the NCI-60 panel. The variant containing intact indole and quinolizine and a methoxy substituent (P18) was most active against a panel of breast cancer cells. We found that the lead compound, P18, effectively inhibits the growth and clonogenic potential of breast cancer cells. Activation of p53 is popularly known to activate pro-apoptotic factors (Puma, Noxa, Bax, etc.) or cell cycle arrest genes (p21, BTG2, etc.) and repress transcription of anti-apoptotic, cell-cycle promoting and survival genes (Vousden and Prives, 2009). Contrary to the prevailing view of the role of induction of apoptosis or growth arrest to mediate tumor suppressor function of p53, several studies have shown tumor suppressor effects of p53 in transgenic mice mutant for cell cycle arrest, senescence and apoptosis (Christophorou et al., 2006; Jiang et al., 2011). Our findings extend beyond cancer cell growth and show the inhibitory effect of p53 activator-P18 on epithelial-mesenchymal-transition, migration, invasion, mammosphere formation and expression of iPSC markers, Oct4, Nanog and Sox2. These studies provide an interesting mechanism involved in biological function of P18 where P18 induces the expression of miR-34a in a p53-dependent manner. We also show that P18 mediates its biological actions via p53 as p53-null cells do not show inhibition of migration, invasion or inhibition of mesenchymal markers in response to P18 treatment.

Our studies offer the first evidence of the ability of sempervirine analog to inhibit EMT and stemness in breast cancer cells. EMT plays an important role in normal physiological processes such as embryonic development, tissue remodeling and wound healing as well as cancer progression by enabling epithelial-derived tumors to acquire mesenchymal characteristics. Cancer cells undergoing mesenchymal transition typically exhibit increased migration and invasion potential as well as stem-like characteristics (Mani et al., 2008; Morel et al., 2008; Santisteban et al., 2009). EMT-related transcription factors act in concert to orchestrate EMT-associated changes including gain of mesenchymal gene expression and the repression of key epithelial genes (Tam and Weinberg, 2013). P18 inhibits the expression of fibronectin and vimentin in p53-wild-type cells along with a gain of E-cadherin expression. Modulation of E-cadherin expression is considered a crucial event in EMT and MET, and EMT-related transcription factors can be grouped as direct or indirect repressors of E-cadherin. Direct repressors of E-cadherin include zinc finger proteins of the SNAIL superfamily, such as SNAI1 (also known as SNAIL), SNAI2 (SLUG) and zinc finger and E-box binding proteins of the ZEB family, such as ZEB1 (TCF8) (Thiery et al., 2009). P18 treatment inhibits the expression of Zeb1, Snail and Slug in p53-wild-type cells. Recently, p53 activity has been associated with the modulation of EMT and stemness features as inactivation of functional p53 induces EMT and associates with stem-like signatures (Spike and Wahl, 2011). While wild-type p53 interacts with MDM2 and Slug to promote degradation of Slug; mutant p53 upregulates the expression of Twist and enhances Slug expression by inhibiting its degradation (Kogan-Sakin et al., 2011).

Tumor suppressor p53 is known to regulate various cellular processes controlling myriad of genes acting as a transcriptional activator as well as repressor either directly or indirectly. The possibility that p53 controls microRNAs allows indirect repression of target gene expression at the posttranscriptional level by p53. The miR-34 family of miRNAs has been recently implicated in the p53 tumor suppressor network (Chang et al., 2007). Wild-type p53 is known to transactivate miR-34 to broadly influence gene expression and promote apoptosis (Chang et al., 2007, 2011; He et al., 2007). We observed that P18 increases miR-34a levels in breast cancer cells. Elevation of miR-34a in primary neuroblastomas and cell lines, inherently expressing low miR-34a expression, inhibits proliferation and activates cell death pathways (Welch et al., 2007). Previous studies have shown that miR-34a is directly regulated by interaction of p53 with consensus p53-binding site in short promoter proximal regions of miR34a (Bommer et al., 2007). p53-mediated activation of miR-34a results in either cell-cycle arrest or senescence or apoptosis depending on the cellular context (He et al., 2007). We observed that P18 inhibits mammosphere formation and miR-34a mimic further enhanced the effect. Tumor suppressor p53 can achieve these numerous biological effects by activating miR-34a and affecting the spectrum of miR-34 regulatory targets in cell-type dependent manner.

Owing to the almost universal alteration of p53 in cancer, several therapeutic approaches to target p53 pathway are under various developmental stages. Many small molecules have been developed to target protein–protein interactions and protein folding pathways involved in p53 regulation. BCl-2 inhibitor ABT-373, RITA, and the MDM2 inhibitors MI-219 and nutlin are good examples of approaches to block protein–protein interactions (Shangary et al., 2008; Vassilev et al., 2004). Small molecules targeting protein folding to reactivate mutant p53 include PRIMA-1 and CP-31398 (Bykov et al., 2002; Foster et al., 1999). Increased expression of MDM2 leading to inhibition of p53 activity in tumors has been the target of development of many small molecules, peptide and aptamer-based therapies. Nutlins were identified from a class of cis-imidazoline compounds (Vassilev et al., 2004). Orally available nutlin analog RG7112 was tested in clinic in liposarcoma patients resulting in stable disease in 14 out of 20 patients and a partial response in one patient but all patients experienced adverse effects such as neutropenia and thrombocytopenia (Ray-Coquard et al., 2012). Orally active, non-imidazoline MDM2 inhibitors RG7388, RO5353 and RO2468 are also under development (Ding et al., 2013; Zhang et al., 2014). Yet another strategy is to block MDM2 and MdmX simultaneously. Small molecule RO-5963 induces the formation of MDM2 and MdmX homo or hetero-dimer that are unable to bind p53 leading to p53 activation (Lee et al., 2011). Reactivation of p53 is a very promising anti-cancer strategy which is currently being tested in clinic. Since a major concern in drug development is the associated side effects, further development of active constitutive agents in natural products and their analogs that have the capability to activate p53 (for example, sempervirine and its analog P18) could provide safe and effective anti-cancer therapeutic strategies.

4.1 Conclusions

Here, we reported the synthesis of novel indolo-pyrido-isoquinolin based alkaloids which were examined for their therapeutic efficacy in breast cancer cells. Our studies showed that one of these compounds, P18, effectively inhibited growth, invasion, migration, EMT and mammosphere formation. Mechanistically, P18 induced p53-miR34a axis to mediate its biological effects on breast cancer cells. These studies provided insight into the specific pathways that are required for P18-mediated inhibition of growth, EMT, invasion, migration and mammosphere-formation and put forth P18 as a rational therapeutic strategy for breast carcinoma.

Conflict of interest

DBA, AN, JET, PR, MJD, BM, NKS, SVM and DS declare no conflict of interest.

Acknowledgments

This work was supported by NIDDK NIH, K01DK077137, R03DK089130 and NCI NIH R21CA185943 (to NKS); NCI NIH R01CA131294, NCI NIH R21CA155686, Avon Foundation, NCI, NIH contract no. HHSN261200800001E (to SVM), Breast Cancer Research Foundation (BCRF) 90047965 (to DS).

Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.molonc.2016.04.003.