Metabolic reprogramming has been recognized as one of the 10 hallmarks of cancer [1
]. Malignant cells need to change their cellular energy metabolism to support unrestrained cell proliferation and to adapt to new microenvironmental conditions and to different challenges. Lipid metabolism is no exception: a sustained biosynthesis of membrane phospholipids is required to meet the demand of rapidly proliferating cells. In fact, alteration in choline (Cho) metabolism has been observed in many cancers [2
] and it has been related to deregulated cell proliferation, invasion, and metastasis. The so-called “cholinic phenotype” consists of increased level of phosphocholine (PCho) and, in general, of total choline-containing metabolites (tCho) mainly due to the overexpression and/or hyperactivation of the α1 isoform of choline kinase (ChoKα1) [9
]. In humans, three isoforms of ChoK have been described: ChoKα1 and ChoKα2 encoded by the CHKA
gene and ChoKβ by CHKB.
In the first step of the Kennedy pathway, these enzymes catalyze the phosphorylation of choline to phosphocholine, ultimately leading to the synthesis of phosphatidylcholine (PtdCho), the most abundant phospholipid of the eukaryotic cell membrane. Although ChoK proteins share high sequence similarity, only the ChoKα1 isoform has been proposed as an oncogenic promoting factor. Increased expression of ChoKα1 has been extensively described in breast cancer, where significantly increased activity has been reported in 40% of patients in correlation with histological tumor grade and poor clinical outcome [5
]. Silencing of CHKA
by RNA interference has been demonstrated to reduce cell proliferation and tumor growth [11
], sensitize cancer cells to chemotherapeutics [13
], and suppress migration and invasion, while CHKB
silencing has no effect [3
]. For these reasons, ChoKα1 has been proposed as a new appealing target for cancer therapy, and during the past decades, extensive efforts have been made to synthesize and improve ChoKα1 inhibitors [14
EB-3D (previously named 10a) is a novel choline-competitive symmetrical biscationic ChoKα inhibitor that was shown to impair cell proliferation in a panel of different cancer cell lines [24
]. Our group recently reported that EB-3D is able to induce apoptosis in T cell acute lymphoblastic leukemia and to synergize with L-asparaginase by targeting the same signaling pathway [8
]. Here, we further investigated the effect of EB-3D-mediated ChoKα inhibition both in vitro and in vivo in breast cancer, where the overexpression and hyperactivation of ChoKα are associated with tumor progression, invasion, and metastasis [13
]. We found that EB-3D, through phosphocholine level reduction, was able to impair cell proliferation, thus triggering cells to senescence via activation of the AMPK-mTOR pathway. Moreover, EB-3D treatment significantly enhanced the antitumorigenic potential of drugs commonly used in breast cancer treatment protocols. Finally, the inhibition of ChoKα by EB-3D reduced cell invasion and migration, with a significant loss of metastatic potential of breast cancer cells in vivo.
Literature studies have deemed ChoKα as a novel cancer target due to its role in survival signaling and in supporting cell proliferation. Moreover, although the oncogenic properties of ChoKα are not fully elucidated, it appears critical for the survival of cancer cells [8
]. The small molecule EB-3D specifically targets the ChoKα enzyme in breast cancer cells, as pointed out by the decrease of PCho and the matched decrease of tCho levels in water-phase extracts of treated cells. We also observed a twofold increase in Cho levels after treatment that can be explained as the accumulation of ChoKα substrate, supporting the enzymatic inhibition. In different breast cancer cell lines, EB-3D induces a strong and irreversible cell proliferation arrest, with the onset of significant cell death only after prolonged time of high dose of drug exposure (20–30-fold GI50
). This behavior has been described for other biscationic symmetrical ChoKα inhibitors [23
], while nonsymmetrical compounds do not affect cell proliferation and viability [18
The lack in PCho biosynthesis has been demonstrated to induce cell growth arrest and accumulation of cells in middle-to-late G1. Moreover, the increase in PCho level is a requirement for the transition through the G1 restriction point [33
] and its reduction is consistent with the reported cell cycle arrest in the G0/G1 phase induced by EB-3D treatment. This late G1 restriction point, known also as cell growth or metabolic check-point, is dependent on mTOR signaling and cyclin E activation. Recent evidence showed that the activation of these signals is controlled by the metabolic sensor AMPK, which is able to suppress the mTOR signal in response to energy stress and nutrient-poor conditions [35
]. Immunoblot data highlighted a modulation of the AMPK-mTOR metabolic pathway induced by EB-3D treatment. Indeed, Trousil et al. recently described that the symmetrical biscationic ChoKα inhibitor ICL-CCIC-0019 reduces mitochondria respiration and ATP production, leading to AMPK activation without increasing reactive oxygen species (ROS) production [23
], revealing a noncanonical mitochondrial damage response [36
]. We report the activation of the AMPK stress sensor that caused the reduction of mTOR phosphorylation and of its downstream targets 4E-BP1, p70S6K, and S6 upon treatment with EB-3D. The dephosphorylated form of 4E-BP1, which sequesters the eukaryotic translation initiation factor 4E (eIF4E), together with the absence of the active hyperphosphorylated form of S6, prevents the initiation and progression of the mRNA translation process. Collectively, these data suggest that ChoKα inhibition by EB-3D is sensed like a metabolic insult in breast cancer cells causing the dephosphorylation of the mTORC1 final effectors required for protein synthesis. These observations corroborate the observed reduction in cell proliferation and G0/G1 cell cycle arrest following EB-3D treatment.
The arrest of cell proliferation is maintained even after compound removal, suggesting that ChoKα inhibition is irreversible or at least that the damages caused by EB-3D are permanent, conversely to what has been reported for the nonsymmetrical ChoKα inhibitor V-11-0711 [18
]. Indeed, to the best of our knowledge, we proved for the first time that cellular senescence can be induced by targeting choline metabolism in breast cancer. The pharmacological inhibition of ChoKα with EB-3D significantly increases the activity of the senescent marker SA-βgal in all tested breast cancer cells and at the same time it is important to note that this phenomenon occurs also in the in vivo E0771-C57BL/6 syngeneic model. In addition, the twofold increase of the GPCho/PCho ratio observed (Figure S1B
) has also been described as a feature of senescent cells, independently from the type of senescence [37
]. The induction of senescence-like phenotype seems to be cancer specific since EB-3D treatment did not induce SA-βgal activity in healthy mammary cells MCF10A and normal human fibroblasts (Figure S5D
The involvement of AMPK signaling in triggering senescence has been already described. In this context, the increased AMP/ATP ratio and AMPK activity were observed during cellular senescence in fibroblasts [38
]. Consistently, sustained AMPK activation was observed during radiation-induced senescence [39
], although other research groups report that activation of AMPK prevents H2
-induced senescence [40
] triggering autophagy [42
]. In this work, we demonstrated that the activation of AMPK by EB-3D appears to drive cellular senescence in breast cancer cells. Indeed, AMPK inhibition by Compound C
reverted EB-3D-induced senescence-like phenotype. Of note, the induction of AMPK is a rapid event, occurring within 18 h of EB-3D treatment, and probably is not directly correlated with the reduction of pCho that does not occur before 24–48 h. Thus, it will be important to further study the possible relationship between the inhibition of ChoKα and the induction of metabolic stress. In this context, Falcon et al. demonstrated different effects on cell viability between ChoKα silencing and pharmacological inhibition by V-11-0711, suggesting a role for the ChoKα protein itself in promoting cancer cell survival that is independent of its catalytic activity [18
Whilst cellular senescence is known to be a permanent and irreversible process, the senescence-associated secretory phenotype (SASP) has been pointed out as a potential strategy to promote tumor progression [43
] and drug resistance [44
]. Thus, the induction of cellular senescence could be a double-edged sword in cancer. Indeed, considering senescence as an irreversible state of growth arrest with the complete loss of proliferation potential, senescence could be perceived as a successful outcome of therapy, since the cells are reproductively “dead” [45
]. Moreover, induced senescence can stimulate immunological response helping the eradication of the tumor cells [46
]. We demonstrated that EB-3D-induced senescence sensitizes breast cancer cells to the apoptotic effect of cisplatin. This result reveals that the onset of cellular senescence in breast cancer cells might be advantageous. EB-3D also enhances the chemotherapeutic effects of 5-FU and doxorubicin, significantly lowering their GI50
values. All together, these data suggest that ChoKα inhibition could be a potential neoadjuvant and could enhance the effects of conventional chemotherapy for breast cancer, although further studies are needed.
In this work, we have also demonstrated the efficacy of EB-3D as a potent antitumor agent in vivo in a syngeneic orthotopic mouse model. It is worth noting that the effect of the drug is evident already after the second administration of 1 mg/Kg intraperitoneally (ip), indicating a potent antitumor effect and favorable pharmacokinetics. Importantly, we did not observe any sign of apparent toxicity.
Since ChoKα is the first enzyme involved in phosphatidylcholine biosynthesis, the most abundant membrane lipid, it is reasonable to assume that ChoKα plays an important role in cell membrane stability and therefore in migration and invasion processes. EB-3D-mediated ChoKα inhibition drastically reduces tumor cell motility and invasiveness in vitro in the highly aggressive MDA-MB-231. Indeed, it has been recently reported that ChoKα inhibition suppresses epithelial-to-mesenchimal (EMT) transition in glioblastoma [47
]. In agreement with Koch et al., we found a deregulation of EMT-related transcription factors of the SNAIL, ZEB, and TWIST families and the canonical E-cadherin/N-cadherin switching. The high mortality rate associated with triple negative breast cancer (the most aggressive) is due primarily to the onset of metastases, mainly targeting lung, liver, bones, and brain. Hence, research efforts should focus also on the development of new therapies for prevention of secondary metastatic lesions. For these reasons, the antimetastatic effect of EB-3D was tested also in vivo using both allogeneic and xenogeneic models. Indeed, an essential component for testing new pharmacological agents is the assessment of their efficacy in preclinical settings. However, very few preclinical models that incorporate the relevant features of human metastatic disease are available. We were able to provide preliminary evidence of reduction of spontaneous lung metastatic nodules in mice treated with EB-3D after primary tumor resection, but the effect became significant when tumor cells were injected intravenously.
4. Materials and Methods
4.1. Cell Lines and Culture Conditions
Human breast adenocarcinoma MCF-7 (ATCC cat n. HTB-22; Manassas, VA, USA) were grown in low-glucose DMEM (PanBiotech, Aidenbach, Germany), MDA-MB-231, and MDA-MB-468 cell lines (provided by Dr. R. Giavazzi, Istituto M. Negri, Milan, Italy and authenticated by STR method) in DMEM (Gibco, Thermo Fisher Scientific, Waltham, MA, USA). Both media were supplemented with 10% fetal bovine serum (FBS), glutamine (2 mM), penicillin (100 U/mL), and streptomycin (100 µg/mL) (all from Thermo Fisher Scientific, Waltham, MA, USA). Mouse breast cancer E0771 cells (provided by Dr. R. Giavazzi, Istituto M. Negri, Milan, Italy and authenticated by STR method) were grown in complete DMEM medium supplemented with 20% FBS. All cell lines were cultured at 37 °C, 5% CO2 for no longer than 15 passages. Mycoplasma testing was periodically performed using Venor GeM OneStep Mycoplasma Detection Kit (Minerva Biolabs, Berlin, Germany). In every experiment, the DMSO concentration never exceeded 0.5%.
4.2. Cell Viability Assay and Drug Combination Sensitivity Assay
The cytotoxic activity of the selected drugs or drug combinations was determined after 72 h of treatment by MTT colorimetric assay (Sigma-Aldrich, Milan, Italy). In drug combination assays, synergism was determined by the calculation of the combination index (CI) using CalcuSyn software (version 2.0, Biosoft) based on the algorithm described by Chou [48
], where synergism is defined by CI < 1, additivity as CI = 1, and antagonism as CI > 1. In additional experiments, the accurate determination of the cell proliferation rate was determined by trypan blue exclusion assay. Cells in exponential growth were treated with EB-3D (time 0) and then collected at the indicated time points. Cells were resuspended in 0.4% trypan blue solution (Thermo Fisher Scientific) and counted on a hemocytometer. Only trypan blue negative cells were considered viable cells. Detailed protocols are available in the Supplementary Materials
4.3. Cell Cycle Distribution Analysis
For flow cytometric analysis of DNA content, 5 × 105 cells in exponential growth were treated with different concentrations of EB-3D for 24 h. Cells were then collected, centrifuged, and fixed with ice-cold ethanol (70%). The cells were then treated with a buffer containing RNAse A (Qiagen, Hilden, Germany) and 0.1% Triton X-100 (Sigma-Aldrich) and then stained with propidium iodide (PI) (Sigma-Aldrich). Samples were analyzed on a Cytomics FC500 flow cytometer (Beckman Coulter, Brea, CA, USA). DNA histograms were analyzed using MultiCycle for Windows (Phoenix Flow Systems).
4.4. Magnetic Resonance Spectroscopy (1H-MRS)
MDA-MB-321 breast cancer cells were seeded and cultured for 24 h in complete growth medium and then treated with EB-3D or DMSO for the indicated time points. Water-soluble extracts were obtained using the dual-phase extraction method. The detailed protocol is in the Supplementary Materials
4.5. Annexin-V/PI Assay
Surface exposure of phosphatidylserine on apoptotic cells was measured by flow cytometry with Cytomics FC500 (Beckman Coulter) by simultaneously adding annexin-V (AV) conjugated to fluorescein isothiocyanate (FITC) and propidium iodide (PI) to cells according to the manufacturer’s instructions (Annexin-V-Fluos staining kit, Roche Diagnostic, Indianapolis, IN, USA).
4.6. C12-FDG Senescence Assay
For flow cytometric analysis of cellular senescence, cells were treated with EB-3D for 72 h and then the medium was replaced with fresh medium containing EB-3D or DMSO for the next 72 h. For rescue experiments, cells were pretreated with 2.5 μM of Compound C
(Santa Cruz Biotechnology, Dallas, TX, USA) for 2 h and then treated with EB-3D for the next 3 days. Senescence was evaluated by flow cytometry on a Cytomics FC500 flow cytometer (Beckman Coulter) using di-β-D-galactopyranoside (C12
-FDG) (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA), a fluorogenic substrate for β-gal activity, as previously described [49
4.7. Western Blot Analysis
Total cellular proteins were extracted from cells using T-PER lysis buffer (Pierce, Milano, Italy) containing phosphatase and protease inhibitors. The protein concentration was determined using the BCA protein assay reagents (Thermo Scientific). Equal amounts of protein were resolved using SDS-PAGE and transferred to PVDF Immobilon-P Membrane (Merck Millipore, Darmstadt, Germany). Membranes were saturated with BSA 3% and probed overnight at 4 °C with specific primary antibodies (listed in the Supplementary Materials
and Methods). Membranes were then washed and incubated with HRP-labeled secondary antibodies (goat anti-rabbit or anti-mouse IgG; Perkin Elmer, Waltham, MA, USA). All membranes were stained using ECL Select (GE Healthcare, Catania, Italy) and visualized with Alliance 9.7 (UVITEC, Cambridge, UK). To ensure equal protein loading, each membrane was reprobed with β-actin antibody (Sigma-Aldrich).
4.8. Scratch-Migration Assay
Nearly confluent MDA-MB-231 cells were gently wounded through the horizontal and vertical axis using a pipette tip. Cells were washed twice to remove cell debris and then treated with EB-3D at the indicated concentration for 48 h. Each time point image was captured at 7× magnification under a stereomicroscope. The distance between the two edges of the scratch was quantified using Adobe Photoshop CS6.
4.9. Cultrex BME Cell Invasion Assay
MDA-MB-231 cells were added to 24-well Transwell inserts of the Cultrex BME (Basal Membrane Extract) Cell Invasion Assay (Trevigen, Gaithersburg, MD, USA) according to the manufacturer’s manual. Invasion was measured 24 h after plating at 485–520 nm using the VICTOR microplate reader (Perkin Elmer).
4.10. Quantitative Real-Time PCR (qRT-PCR)
Total RNA was extracted with TRIzol reagent (Invitrogen) according to the manufacturer’s protocol and RNA purity and concentration were determined by measuring the spectrophotometric absorption at 260 and 280 nm on NanoDrop ND-1000. One microgram of RNA was reverse-transcribed into first strand cDNA using Superscript III Reverse Transcriptase (Life technologies, Paisley, UK) and random primers following the manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) reaction was carried with SYBR Green PCR Master Mix (Life Technologies) with ABI 7900 system (Applied Biosystems, Foster City, CA, USA) using specific primers listed in the Supplementary Materials
and Methods. Each reaction was performed in triplicate and mRNA levels of target genes were normalized by the housekeeping gene GUS
and expressed as a fold change relative to control using the 2−ΔΔCt
method. Data are represented as mean ± standard error of the mean (SEM) of three independent experiments.
4.11. In Vivo Tumor Growth
Animal experiments were approved by the local animal ethics committee (OPBA, Organismo Preposto al Benessere degli Animali, Università degli Studi di Brescia, Italy) and were performed in accordance with national guidelines and regulations. Procedures involving animals and their care conformed with institutional guidelines that comply with national and international laws and policies (EEC Council Directive 86/609, OJ L 358, 12 December 1987) and with “ARRIVE” guidelines (Animals in Research Reporting In Vivo Experiments).
Seven-week-old C57BL/6 female mice were orthotopically injected into the mammary fat pad with 4 × 105 E0771 mammary carcinoma cells. When tumors were palpable, mice were randomized to control and treated groups. Treatment was performed every other day by intraperitoneal (ip) injection of EB-3D (1 mg/kg) or vehicle (DMSO) in 100 µL final volume. Tumors were measured in vivo using a calipers:tumor volume V (mm3) that was calculated according to the formula V = (D × d2)/2, where D and d are the major and minor perpendicular tumor diameters in mm, respectively.
Tumor volume data were analyzed with a two-way analysis of variance, and individual group comparisons were evaluated by the Bonferroni correction.
At the end of the experimental procedure, tumors were harvested, weighted, photographed, and embedded in OCT and frozen for histological processing.
4.12. In Vivo Metastasis Models
Animal experiments were approved by the local animal ethics committee (OPBA, Organismo Preposto al Benessere degli Animali, Università degli Studi di Brescia, Italy) and were executed in accordance with national guidelines and regulations. Procedures involving animals and their care were conformed with institutional guidelines that comply with national and international laws and policies (EEC Council Directive 86/609, OJ L 358, 12 December 1987) and with “ARRIVE” guidelines (Animals in Research Reporting In Vivo Experiments).
For spontaneous metastases, 5 × 105 E0771 cells were injected into the mammary fat pad of C57BL/6 female mice and tumors were resected when they reached the 8 × 13 mm size. After one week of recovery, mice were treated ip every other day for four weeks with vehicle or EB-3D (2.5 mg/kg) in 100 µL final volume.
For experimentally induced metastasis, murine E0771 (2 × 105) or human MDA-MB-231 (8 × 105) breast cancer cells in 100 µL of PBS were injected intravenously into the tail vein of 7-week-old C57BL/6 or NOD/SCID female mice, respectively. Animals were treated every other day by ip injection with EB-3D (2.5 mg/kg) or vehicle (DMSO) in 100 µL final volume for 3 (for E0771 cells) or 7 weeks (for MDA-MB-231 cells).
At the end of the experimental procedures, mice were sacrificed, lungs were harvested, weighted, formalin-fixed, and macrometastases were counted under a dissecting microscope. Lungs were then embedded in paraffin for histological processing and for the counting of micrometastases. Hematoxylin and eosin (H&E) staining was performed on five sections/specimens and lesions containing 50–200 cells were considered as micrometastases.
4.13. Histological Analyses
OCT-embedded E0771 tumors were sectioned and stained with anti-Ki67 antibody or with x-gal using Senescence Cells Histochemical Staining Kit (Sigma-Aldrich) according to the manufacturer’s instructions. Nuclei were counterstained by Meyer’s Hematoxylin.
Formalin-fixed, paraffin-embedded samples were sectioned, dewaxed, hydrated, and stained by H&E. All images were captured by a video-confocal microscope (T1 microscope, Zeiss, Oberkochen, Germany) using a 10× objective. Images were compiled for figures using Adobe Illustrator (Adobe Systems Inc., San Jose, CA, USA).
4.14. Statistical Analyses
Graphs and statistical analyses were performed using GraphPad Prism software (GraphPad, La Jolla, CA, USA). All data in graphs represented the mean of at least three independent experiments ± SEM. Statistical significance was determined using Student’s t-test or ANOVA (one- or two-way) depending on the type of data. For multiple test comparison, Bonferroni or Newman–Keuls corrections were applied. Asterisks indicate a significant difference between the treated and the control group, unless otherwise specified. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.